University of Nebraska - Lincoln
DigitalCommons@University of Nebraska - Lincoln
P. D. Harms Publications
Management Department
Spring 2015
How Functionalist and Process Approaches to
Behavior Can Explain Trait Covariation
Dustin Wood
Wake Forest University, [email protected]
Molly Hensler Gardner
University of Alabama at Birmingham
Peter D. Harms
University of Nebraska - Lincoln, [email protected]
Follow this and additional works at: http://digitalcommons.unl.edu/pdharms
Part of the Business Administration, Management, and Operations Commons, Human
Resources Management Commons, and the Organizational Behavior and Theory Commons
Wood, Dustin; Gardner, Molly Hensler; and Harms, Peter D., "How Functionalist and Process Approaches to Behavior Can Explain
Trait Covariation" (2015). P. D. Harms Publications. Paper 8.
http://digitalcommons.unl.edu/pdharms/8
This Article is brought to you for free and open access by the Management Department at DigitalCommons@University of Nebraska - Lincoln. It has
been accepted for inclusion in P. D. Harms Publications by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln.
Published in Psychological Review 122:1 (2015), pp. 84-111; doi: 10.1037/a0038423
Copyright © 2015 American Psychological Association. Used by permission.
“This article may not exactly replicate the final version published in the APA journal. It is not the copy of record.”
Submitted July 15, 2013; revised October 15, 2014; accepted October 17, 2014.
digitalcommons.unl.edu
How Functionalist and Process Approaches to Behavior
Can Explain Trait Covariation
Dustin Wood, Department of Psychology, Wake Forest University
Molly Hensler Gardner, Department of Psychology, University of Alabama at Birmingham
P. D. Harms, Department of Management, University of Nebraska–Lincoln
Abstract
Factors identified in investigations of trait structure (e.g., the Big Five) are sometimes understood as explanations or
sources of the covariation of distinct behavioral traits, as when extraversion is suggested to underlie the covariation of
assertiveness and sociability. Here, we detail how trait covariation can alternatively be understood as arising from units
common to functionalist and process frameworks, such as self-efficacies, expectancies, values, and goals. Specifically,
the expected covariation between two behavioral traits should be increased when a specific process variable tends to
indicate the functionality of both traits simultaneously. In 2 empirical illustrations, we identify a wide array of specific
process variables associated with several Big Five-related behavioral traits simultaneously, and which are thus likely
sources of their covariation. Many of these, such as positive interpersonal expectancies, self-regulatory skills, and preference for order, relate similarly to a broad range of trait perceptions in both studies, and across both self- and peerreports. We also illustrate how this understanding of trait covariation provides a somewhat novel explanation of why
some traits are uncorrelated. As we discuss, a functionalist understanding of trait covariation as arising through functionalist or process variables has implications for many basic issues in personality psychology, such as how personality traits should be measured, mechanisms for personality stability and change, and the nature of personality traits
more generally.
Keywords: Personality traits, functionalism, expectancy-value models, social cognition, covariation
Supplemental materials are included following the References.
Why do individual differences in levels of traits such as
assertiveness and sociability covary? Frequently, personality psychologists have answered this by appealing to broad
factors identified by investigations of trait structure, such as
the extraversion, agreeableness, conscientiousness, neuroticism, and intellect/openness factors that constitute the Big
Five and Five Factor Model (FFM) frameworks (Goldberg,
1993; McCrae & Costa, 2008), or the similar HEXACO factors (Ashton & Lee, 2007). Factors of this sort have sometimes
been considered “source traits” or “basic tendencies” due to
the assumption that they show some approximation to the
most important sources “underlying” the more specific traits
used to identify them (Cattell, 1950; McCrae & Costa, 1995).
Here, we will describe how trait covariation can be accounted for by functionalist or process approaches to behavior. Such frameworks understand behaviors as being
means toward desired ends, and enlist constructs such as
self-efficacies, expectancies, and values as preferred explanatory units (e.g., Ajzen, 1991; Bandura, 1977; Feather, 1982;
Heckhausen, 1977; Mischel & Shoda, 1995; Vroom, 1964).
Functionalist and process frameworks have typically been
applied to understanding between-person or within-person
variation of a single behavioral tendency; however, they
have not typically been applied to understanding trait covariation, which is among the central phenomenon structural factors are enlisted to help explain (Ashton & Lee, 2005;
Tellegen, 1991).
Here, we will first briefly describe how structural approaches have traditionally been enlisted to account for trait
covariation. We then describe how functionalist and process
frameworks tend to explain behavioral trait levels, and how
this understanding can be extended to explanations of trait
covariation without invoking structural factors in an explanatory fashion. We illustrate this approach to trait covariation
Acknowledgements — We thank Kate Rogers and Jessica Wortman for their help with their assistance in the generation of the items used in Study
1; Lew Goldberg, Maureen Barckley, and Chris Arthun of the Oregon Research Institute for their help in identifying resources collected with the Eugene-Springfield Community Sample for the analyses in Study 2; and Sarah Thomas and Seth Spain for their help in developing the IPIP item categorizations used in Study 2. We would also like to thank Jaap Denissen, Grant Edmonds, William Fleeson, David Funder, Lewis Goldberg, Eranda
Jayawickreme, Lara Kammrath, Brent Roberts, Ryne Sherman, Mary Jane Skelly, Jenn Lodi-Smith, Seth Spain, Sanjay Srivastava, Simine Vazire, and
Christian Waugh for their helpful comments on a draft of this article.
84
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation 85
Figure 1. Conceptualizations of sources of trait covariation consistent with structural equation models (Figure 1A) and functionalist
frameworks (Figure 1B). Circles represent factors identified by structural analyses, boxes represent measured variables. Solid lines
indicate positive effects and dashed lines indicate negative effects. In Figure 1A, structural factors are considered underlying factors
of the measured traits; in Figure 1B, structural factors are considered composite averages (“components”) of the measured traits.
through two empirical illustrations. As we will show, this results in a picture of the sources of trait covariation that looks
quite different from the picture provided by structural factors. Finally, we will discuss some ways in which this functionalist approach to covariation might be reconciled with
structural approaches. Given the considerable role of structural factors such as the Big Five in contemporary personality psychology, we also outline some of the broader implications of a functional model of trait covariation.
Explanation of Behavioral Trait Covariation Through
Broad Structural Factors
Beginning with Cattell’s (1946, 1950) work, structural approaches to the covariation of behavioral traits have regularly proceeded by using factor analysis to identify factors
(e.g., extraversion) which, in conjunction with factor loadings, can roughly reproduce the observed interrelationships
between distinct behavioral traits (e.g., sociability, assertiveness, positive affectivity). In the simplest case where we are
examining the correlation between two variables, factor analysis can always extract a single factor with factor loadings
which will be precisely equal to the square root of the correlation between the two variables. For instance, as shown in
Figure 1A, if we were interested in identifying a factor underlying the covariation between assertiveness and sociability, and these two traits show an association of r = .49,
we can use factor analysis to identify a factor which has .70
loadings with both traits. These loadings mathematically
serve as a way to externalize the correlation between the
variables into a structural factor. In this example, by multiplying the .70 loadings with the factor together, we can say
that assertiveness and sociability have an indirect .49 correlation “through” the common factor, which we might label
“extraversion.”
When we are looking at the association of two variables,
factor analysis and other structural techniques can always
identify a single factor which can recover the association between variables perfectly. As the number of items in the analysis increases, a small number of factors will no longer be
able to reproduce the correlations between items perfectly but
may nonetheless be able to reproduce the original matrix quite
well. Similarly, with more than two items the empirical factor
estimates given to individuals will shift from a straight averaging of the variables to a weighted average where items are
weighted by their factor loadings.
There are a number of ways we might interpret the variables that result from analyses of trait structure. However, a
86 Wood, Hensler Gardner, & Harms
particularly influential perspective has been the understanding that these may approximate the major causes of covariation among diverse traits. Cattell (1946) regularly discussed
a factor resulting from these analyses as potentially approximating a “source trait,” which can be thought of as a “cause
or determiner of several trait-elements showing covariation”
(Cattell, 1950, p. 33), and a “possible explanation of how the
actually existing forms may have originated” (Cattell, 1946,
p. 79). We believe this is the most intuitive way to interpret the usual graphical representation of structural factors,
which depict them as having causal effects on the narrower
traits in the analysis, as shown in Figure 1A.
In more contemporary personality psychology, this interpretation of structural factors most clearly approximates the
view traditionally described within the Five Factor Theory
of McCrae and Costa (1995), where the Five Factors (which
largely approximate the Big Five factors identified in lexical studies; Goldberg, 1993) are understood as “underlying
tendencies [that] cause and thus explain (in general and in
part) the consistent patterns of thoughts, feelings, and actions
that one sees” (p. 236). However, other personality investigators and methodologists sometimes discuss structural factors as approximating “underlying” constructs which “impact” the narrower traits used to identify them (e.g., Caspi et
al., 2014; DeYoung, Quilty, & Peterson, 2007; Lewis & Bates,
2014; Roberts, Chernyshenko, Stark, & Goldberg, 2005).
There are a range of views as to the meaning of structural factors. Many personality investigators see such factors as providing a means to summarize and organize a person’s standing on diverse traits. For instance, Eysenck (1970)
suggests “we would be wrong in saying that a person behaves in a sociable fashion because he has the trait of sociability, but then trait psychologists . . . [generally] postulate
the trait in question simply as a descriptive variable” (p. 26),
and similarly, Saucier and Goldberg (2001) note that analyses of trait structure “can provide a framework for description, but not necessarily for explanation” (p. 848). From this
perspective, an “extraversion” factor might be best regarded
as a summary rather than as approximating a cause of tendencies to be sociable, assertive, happy, sensation-seeking
(e.g., John, Naumann, & Soto, 2008; Ozer & Reise, 1994; Wiggins, 1997; Wood, in press). As detailed in Figure 1B, this understands an individual’s standing on structural factors as
caused by their standing on narrow factors (by means of a
mathematical averaging) and not vice versa: the causal arrows should point from the specific traits toward the structural factor. However, interpreting structural factors in
this manner means that structural analyses may not get us
much closer to understanding the sources of trait covariation. We are more or less back to square one: If extraversion
is a summary rather than a common cause of the covarying
in
Psychological Review 122 (2015)
tendencies to be sociable and assertive, than what causes
these tendencies to covary?
Functionalist Approaches to Trait Levels and Their
Covariation
We argue that trait covariation can be understood by
an extension of how many functionalist or process approaches already explain trait variation. Consequently, we
will first discuss how behavioral trait levels are considered
to be shaped within a range of functionalist and process approaches, and then detail additional principles useful for extending this to the problem of trait covariation.
First, it is useful to describe how we will be using the term
“trait.” There is a frequent tendency for researchers to equate
traits with factors identified in studies of trait structure, as
when investigators ask “How are values and goals associated
with traits?” by correlating measures of such motivational
units with a Big Five measure (e.g., Roberts & Robins, 2000;
Roccas, Sagiv, Schwartz, & Knafo, 2002). However, we wish
to clearly distinguish between these terms. First, we will refer to structural factors as variables such as the Big Five or
HEXACO factors that are identified by statistical procedures
such as factor analysis. In contrast, we will use the term trait
in a manner somewhat closer to Guilford’s (1959) definition
of a trait as “any distinguishable, relatively enduring way
in which one individual differs from others” (p. 6; Johnson,
1999; Roberts, 2009). That is, a trait may range from an individual’s tendency to be tall, male, right-handed, well-liked,
sociable, trusting, like vanilla ice cream, or have a high testosterone level. Given the very wide range of traits individuals can have, it is useful to identify conceptually important trait classes, and how they should relate to one another.
We use the term behavioral traits to be essentially synonymous with behavioral tendencies. This largely parallels models which conceptualize an individual’s level of a behavioral
trait as the individual’s observed or expected likelihood of
performing a certain class of behaviors (Buss & Craik, 1983;
Fleeson, 2001; Wiggins, 1997). From a functional perspective,
behaviors are generally understood as means to ends. For instance, Almlund, Duckworth, Heckman, and Kautz (2011)
consider an individual’s personality system to be “a strategy
function for responding to life situations” (p. 8). Similar assumptions are represented in a wide range of social–cognitive, self-regulatory, evolutionary, and economic models of
behavior (e.g., Ajzen, 1991; Carver & Scheier, 1998; Fleeson
& Jolley, 2006; Gintis, 2007; Kenrick et al., 2009; Kruglanski
et al., 2002; Minsky, 2007). We will refer to the idea that an
individual’s behavioral trait levels are shaped, in large part,
by their real or perceived functionality toward achieving the
individual’s desired ends as the functionality assumption.
1. Note that factor loadings can be used more effectively to reproduce the correlation matrix when using principal axis factoring (PAF) than principal component analysis (PCA). This is because factor analysis by PAF will summarize only the interitem correlations, whereas PCA will additionally base factor loadings on item variances (i.e., the diagonals of the correlation matrix). If an individual’s level of a factor (e.g., extraversion)
is considered simply a summary of the observed variables (e.g., extraversion equals the mean of the individual’s estimated sociability and assertiveness, as represented in Figure 1B), the correlation between the individual’s standing on the structural factor and the original items will more
closely approximated by the PCA loadings (in this example: approximately .87) than by the PAF loadings. Technically, when the variables identified in a factor analysis is considered as “unobserved hypothetical variables that underlie and explain the observed correlations” they are regarded as factors and if considered as “linear composites of observed variables” they are regarded as components or principal components (Loehlin,
2004, pp. 28–29). For simplicity however, we will refer to variables such as the Big Five or HEXACO dimensions as structural factors regardless
of whether they are understood as causes or summaries of covarying traits.
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation Functionality Indicators
Several models suggest that a behavior’s functionality to
the actor can be understood as being established by the operation of three additional broad trait classes; see especially
Ajzen (1991), Bandura (1977), Feather (1982), Gintis (2007),
Heckhausen (1977), Tolman (1938), and Vroom (1964). Specifically, the functionality of a behavior to an actor can be
understood as being established by (a) ability/efficacy traits,
which concern the individual’s expected ease of performing
trait-identifying behaviors; (b) expectancy traits, which concern the expected outcomes of an individual’s performance
of such behaviors; and (c) valuation traits, which indicate the
individual’s expected valuation of particular outcomes. We
will refer to more specific traits that can be classified into
these three trait classes collectively as functionality indicators,
or FIs, to indicate that they can be understood to influence
levels of a behavioral trait primarily by indicating or shaping the behavior’s functionality to the actor. Specifically, an
individual should have a higher level of a behavioral trait if
he or she (a) tends to find the associated behaviors easy to
do; (b) tends to expect them to increase the likelihood of particular outcomes (e.g., attention, power, acceptance); and (c)
tends to value those outcomes.
These units also correspond highly with major classes
found in a range of social–cognitive and process models of
behavior (e.g., Dweck & Leggett, 1988; Fleeson & Jolley, 2006;
Mischel & Shoda, 1995). A theme we will revisit is that individuals’ own explanations for their high or low levels of a
behavioral trait regularly enlist FIs. These three functionalist
trait classes can thus be used to organize or classify different
forms of explanations regarding why individuals have the
traits that they do. We continue by elaborating on the nature
of these trait classes, how they are linked to units in other
prominent models of behavior, and their role in influencing
behavioral trait levels.
Abilities/efficacies. If we understand an individual’s
behavior as means to desired ends, then abilities and efficacies concern which means the individual has available.
Abilities and efficacies are largely analogous to units such
as intention-behavior or efficacy expectancies (Bandura, 1977;
Heckhausen, 1977), and affordances, competencies, skills, capabilities, constraints, and perceived behavioral control described
in other frameworks (Ajzen, 1991; Almlund, Duckworth,
Heckman, & Kautz, 2011; Dweck & Leggett, 1988; Gibson,
1977; Gintis, 2007). Units such as resources or self-regulatory
plans/scripts (Fleeson & Jolley, 2006; Mischel & Shoda, 1995)
can be thought of as linking to behavioral traits in large
part by impacting more general abilities to perform associated behaviors.
Within these and several other frameworks, individuals
are assumed to operate by a law of least effort, choosing behaviors that require less expenditure of effort or resources
(Brehm & Self, 1989; Kahneman, 2011; Thorndike, 1913). For
instance, Ramona and Beatrice may both try to keep their
rooms clean, but Ramona has an easy time doing so while
Beatrice experiences this as more difficult and effortful; this
difference in expected ease of performance will likely result
in Ramona having a cleaner room. Ability traits do not concern specifically how or why the person finds an action easy
or hard to perform. Ramona might find it easier to keep her
room clean due to having better organizational skills, having
87
more energy or better cleaning supplies, or for other reasons.
Abilities simply indicate that the actor somehow finds the
behaviors associated with the behavioral trait easier to perform. An individual’s abilities and efficacies are indicated by
explanations that begin with phrases such as “I [can/can’t]
. . .” or “I find it [easy/difficult] to . . .”
Expectancies. Whereas abilities concern the expected
ease of performing a behavior, expectancies concern how
the environment is expected to change or react after the behavior has been performed. Expectancies in this manner
are most directly analogous to behavior-outcome expectancies (Bandura, 1977; Heckhausen, 1977), and norms and contingencies of reinforcement in other frameworks (Ajzen, 1991;
Cialdini & Trost, 1998; Skinner, 1965).
Within these and many other models, an individual’s
behavioral tendencies are expected to follow the law of effect: increasing when they tend to have desirable or useful effects on the environment, and decreasing when they
do not (Skinner, 1981; Thorndike, 1913). To alter our previous example slightly, perhaps Ramona and Beatrice are
equally capable of keeping their room clean, but it is the
consequences of keeping their rooms clean that differ: Ramona finds having a clean room leads to greater approval
from her parents, while Beatrice does not. This will again
likely result in Ramona having a cleaner room, due now to
differences in this behavior’s expected effects. More generally, although many individuals may be equally afforded
the opportunity to behave in a certain way, they may nonetheless differ in their behavioral traits due to differences in
how their environment reacts to these behaviors. Such expectancies are indicated from explanations of behavior that
begin with phrases such as “I [believe/feel/expect] that . .
.” and “I find [doing X] to . . .”
Valuations. Although different individuals may experience the same outcomes after performing a behavior, they
may differ in the extent to which the outcomes are experienced as desirable and valuable, or useful to other ends.
Whereas expectancy traits concern relations between behaviors and outcomes, valuation traits concern what outcomes the actor finds desirable, or which ends they are trying to maximize. Valuation traits have been referred to as
instrumentalities (Feather, 1982; Vroom, 1964) in other expectancy-value frameworks, and as motives, needs, values,
preferences, and contingencies of utility in other frameworks
(Almlund et al., 2011; Atkinson, 1957; Gintis, 2007; Mischel
& Shoda, 1995; Schwartz, 1992). Goals serve an analogous
function in a range of social–cognitive and self-regulatory
frameworks (Carver & Scheier, 1998; Dweck & Leggett,
1988; Fleeson & Jolley, 2006; Kruglanski et al., 2002) in that
they specify the ends a person would like to attain.
Broadly, valuation traits should impact behavior by specifying the outcomes individuals are trying to promote. To alter our example a final time, perhaps both Ramona and Beatrice can keep their rooms clean with equal ease, and both
receive equal approval from their parents if their room is
clean. However, Ramona experiences her parents’ approval
as rewarding and desirable, whereas Beatrice does not. This
should again result in Ramona having a cleaner room. Valuation traits are indicated from explanations of behavior that
begin with phrases such as “I [like/dislike] . . .”, I [value/
care about] . . .”, and “I typically [try/want] to . . .”
88 Wood, Hensler Gardner, & Harms
Explaining Behavioral Trait Levels by Functionality
Indicators
Considered collectively, behaviors which an individual
finds easy to perform and which are expected to result in
effects the individual finds valuable can be understood as
having higher functionality, utility, or expected value (Atkinson, 1957; Eccles & Wigfield, 2002; Feather, 1959; Gintis, 2007;
Tolman, 1938). In turn, such behaviors should have greater
motivational force (Vroom, 1964), and become the individual’s behavioral intentions (Ajzen, 1991). Through this process,
behaviors with high real or perceived functionality to the actor should be expected to increase in frequency, and consequently behavioral trait levels should come to be correlated
with their functionality to the actor. There are a couple points
worth briefly noting concerning this manner of explanation.
First, it is not necessary to posit individuals as consciously
calculating the functionality of every behavior they perform
to understand behavioral traits as shaped in this manner.
Considerations of whether the behavior is more desirable
than alternatives may often be made in a very quick manner outside of conscious awareness (Gigerenzer & Goldstein,
1996; Kahneman, 2011). Behavioral tendencies can also be
effectively shaped by their functionality by simple learning processes rather than by conscious deliberation (Skinner, 1981; Thorndike, 1913). An individual’s current levels
of behavioral traits can frequently be understood as being
shaped by their past functionality, and then operate principally as habits (Allport, 1937; James, 1890; W. Wood & Neal,
2007). This can lead current behavioral trait levels to depart
from their current functionality.
Second, as noted by McCabe and Fleeson (2012), the shaping of behavior by process variables should be understood
as operating first and foremost at the within-person level of
analysis. For instance, individuals will tend to act extraverted
in moments when they want to promote extraverted-related
ends (e.g., make friends, convey information, entertain others), and more introverted when they don’t. However, as
they also note, individuals also differ in how much they typically desire such ends, and consequently functional explanations should be able to be generalized to explaining not
just within-person variation, but much between-person variation in behavioral trait levels as well.
Third, a given behavior should be understood as having a
wide range of effects on the environment (Bertalanffy, 1968;
Katz & Shapiro, 1985; Kruglanski et al., 2002). For instance,
behaving in the manner we label responsible may tend to increase the fulfillment of commitments to others, interpersonal relatedness, status, effort expenditure, among many
other things. In turn, functional frameworks regard behavioral trait levels as being considerably shaped by the functionality of their effects. In an important sense, the behavior’s
effects are the behavior’s causes: through various mechanisms, behaviors with the more functional effects are selected
over alternatives (Skinner, 1981; Tooby & Cosmides, 1990).
This leads fairly directly to the expectation that functional explanations of behavioral trait levels should be fairly
complex. The overall functionality which shapes behavioral trait levels is itself conceptualized as representing the
sum of probabilities that a behavior will affect diverse outcomes, weighted by the actor’s valuation of each outcome
(Feather, 1982; Fishbein, & Ajzen, 1975; Gintis, 2009). Consequently, we should be able to influence an individual’s
in
Psychological Review 122 (2015)
level of a behavioral trait by influencing a wide range of FIs.
For example, we may be able to increase an actor’s level of
responsible behavior by making the actor care more about
others impacted by such actions, making “irresponsible” alternatives less tempting or less available, increasing the link
between responsible action and long-term goal attainment
(e.g., raises, promotions), making it easier to behave responsibly, or by several other routes (Magidson, Roberts, Collado-Rodriguez, & Lejuez, 2012; Wood, Larson, & Brown,
2009). This can be understood more generally as the principle that there are usually multiple means to a given end
(Austin & Vancouver, 1996; Bertalanffy, 1968; Kruglanski et
al., 2002), and parallels social–cognitive explorations of the
diverse processes that influence a specific behavioral trait
(e.g., Dweck & Leggett, 1988; McCabe & Fleeson, 2012; Metcalfe & Mischel, 1999; Morf & Rhodewalt, 2001).
Explaining Behavioral Trait Covariation by
Functionality Indicators
We have just described that a particular behavioral trait
should have diverse effects. However, any particular trait
should be understood as potentially affecting many other
traits simultaneously. Particularly important to our functionalist explanation of trait covariation is the fact that any particular FI or process variable may affect the functionality of
several distinct behavioral traits simultaneously (e.g., Lukaszewski, 2013).
Mathematically, when one of the various FIs that affects
one behavioral trait also affects a distinct behavioral trait,
we can multiply this FI’s correlations with the two traits together to estimate the expected correlation between the two
traits “through” the FI. For instance, the tendency to see others positively has been found to show moderate correlations
with traits in all Big Five trait domains (Wood, Harms, &
Vazire, 2010), which suggests that seeing others positively
might tend to increase levels of kindness, sociability, assertiveness, responsibility, and so on. Because multiple traits are
impacted by a common FI in this example, this should result
in some level of covariation between these traits. As depicted
hypothetically in Figure 1B, if perceiving others positively
has a .30 effect on assertiveness and a .40 effect on sociability, this should result in a .30 × .40 = .12 increase in the correlation between these two traits “through” the tendency to
see others positively.
More generally, when a particular FI tends to influence
two behaviors in the same direction simultaneously, this
will serve to increase the correlation between the behavioral
traits, and when a particular FI tends to increase levels of
one behavioral trait but decrease levels of another, this will
serve to decrease the correlation between the two traits (i.e.,
make the correlation more negative). Note that this is largely
the same way that structural factors are enlisted to explain
covariation: By comparing Figures 1A and 1B, we see that
structural factors such as extraversion and FIs such as “perceiving others positively” are both cast as common causes of
diverse behavioral traits, and thus sources of covariation.
Differences in structural and functional explanations of
correlated traits. However, beyond this similarity there are
some important differences. One concerns the nature of the
explanatory variable. A structural factor is not measured separately from the traits it is posited to influence, but rather is
generally “inferred” from their covariation. It is this concern
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation that has led some to regard such explanations of covariation as
circular (e.g., Cervone, 1999; Mischel, 1968). In contrast, there
are various reasons to think that explanations of behavioral
trait covariation by FIs are less prone to circularity. For instance, when we find a tendency to see others positively (a
type of expectancy) to relate to behavioral traits in several different Big Five domains (Wood, Harms, et al., 2010), it is not
difficult to see functional reasons for these relationships. Individuals who expect others to have negative traits may sensibly
see little value in meeting and socializing with others (acting
extraverted and open), being nice to others (acting agreeably),
and completing their commitments (acting conscientious), and
adjust their behavioral tendencies accordingly.
A second important difference concerns the number of
variables that will be necessary to explain covariation. Controlling for an FI will tend to reduce the association of the
two variables toward zero, but because several FIs should
influence levels of a given behavioral trait independently of
one another, several distinct FIs should typically be necessary to fully explain the correlation between two variables,
analogous to multiple mediation. For instance, we may find
that positive perceptions of others, social skills, enjoyment
of attention, disliking solitude, and other distinct FIs may be
necessary to fully explain the covariation between assertiveness and sociability. Whereas structural factor models may
explain the covariation of several distinct behavioral traits
through a single structural factor, a functionalist explanation
of covariation suggests that there will generally be several
distinct FIs that must be identified to understand why even
two behavioral traits covary at the level they do.
Differences in structural and functional explanations of
uncorrelated traits. Structural investigations frequently place
a considerable emphasis on identifying orthogonal factors.
This effort is motivated in part by the understanding that
correlated traits have common causes, while uncorrelated
factors may not (Ashton, Lee, Goldberg, & de Vries, 2009;
DeYoung, 2006). However, given that functionalist frameworks construe levels of any particular behavioral trait as
being shaped by many distinct FIs or process variables (e.g.,
Cramer et al., 2012; Fleeson & Jolley, 2006; Kruglanski et al.,
2002; Mischel & Shoda, 1995), it may actually be more surprising to find that uncorrelated behavioral traits share none
of these in common. Instead, we expect that uncorrelated behavioral traits will generally share some common FIs, but
that the FIs that increase the correlation between the two
traits are equal in strength to the FIs that decrease the correlation, and thus pull for correlations between the two traits
in opposite directions which suppress one another (mathematically cancel out; MacKinnon, Krull, & Lockwood, 2000).
To illustrate, as shown in Figure 1B, we might observe
that tendencies to be assertive and polite have an estimated
correlation near r = 0. However, if we were to examine the
various FIs which influence an individual’s levels of assertive and polite behavior, we might find that these two tendencies do in fact share some FIs that influence both traits simultaneously. For instance, desiring power likely increases
assertiveness but decreases politeness (Harms, Roberts, &
Wood, 2007; Roberts, O’Donnell, & Robins, 2004), while perceiving others positively likely increases both assertiveness
and politeness (Wood, Harms, et al., 2010). These and other
FIs may suppress one another, resulting in a negligible correlation between the two traits. Thus, an observed correlation
89
near zero between two behavioral traits may regularly mask
the fact that various specific FIs or process variables are common influences on both traits.
Two Empirical Illustrations
Although some investigators have begun to provide empirical accounts of how diverse behavioral traits central to
the Big Five may be shaped by a common FI (e.g., Lukaszewski, 2013; Wood, Harms, et al., 2010), past empirical research has typically only examined one or a few FIs that may
promote a range of behaviors or self-perceptions associated
with Big Five traits within a single investigation. We thus
conducted two studies to identify more comprehensively the
major FIs that may shape levels of Big Five-related behavioral traits and their covariation. We aimed to identify specific FIs which may be common causes of traits within the
same Big Five domain (e.g., sociability and assertiveness), of
correlated traits in different Big Five domains (e.g., sociability and politeness), and of negligibly correlated traits (e.g.,
orderliness and anxiety).
For Study 1, we adapted a qualitative-to-quantitative approach frequently utilized by Buss and colleagues (e.g., Buss,
Gomes, Higgins, & Lauterbach, 1987; Kyl-Heku & Buss, 1996;
Meston & Buss, 2007; see also Yang et al., 2014) where many
reasons for performing trait-related behaviors are first generated through a qualitative method, and then transformed
into questionnaire items and empirically evaluated as correlates of these behavioral traits in a new sample. This method
was adopted in large part because, as noted by Johnson
(1999), people spontaneously explain actions by “suggesting that the actor (a) desired a certain goal, (b) believed that
the action would help him/her reach the goal, and (c) had
the ability to perform the act” (p. 449; Malle, 1999). That is,
the reasons people give can typically be classified as FIs.
For Study 2, we aimed to replicate the major contours
of Study 1 by utilizing the massive International Personality Item Pool (IPIP) and other resources collected within the
Eugene-Springfield Community Sample (ESCS) to replicate
major features of the FIs that may underlie trait covariation
identified in Study 1. Because these participants were also
rated by close others, we were afforded the opportunity to
examine how FIs were associated not just with self-perceptions, but with how participants were perceived by others.
Measurement Strategy
Many of the FIs that promote behavioral traits and trait
perceptions central to the Big Five are certain to be indicated
by items already found in existing personality trait measures.
For instance, people regularly explain that they act sociably
because they enjoy social interactions, and items concerning
enjoyment of social interactions can be found within common
Extraversion scales. A functionalist framework can be used as
a classification framework which suggests that items referring
to FIs (e.g., enjoying social interactions; having social skills)
and to behavioral tendencies (e.g., socializing) should be classified into separate categories even if highly correlated. In essence, placing such items on a single scale confounds functional causes with their behavioral effects. Consequently, we
separate items referring to behavioral tendencies and abstract
trait perceptions from items referring fairly explicitly to abilities/efficacies, expectancies, and valuations.
90 Wood, Hensler Gardner, & Harms
An important reason for doing this is because we are interested in documenting the wide range of distinct behavioral traits and self-perceptions that might be associated with
a single FI. Indeed, this is central to our functionalist explanation of trait covariation. Separating the measurement of FIs
from behavioral traits affords the ability to identify whether
a particular FI is associated with behavioral traits across multiple trait domains simultaneously. For instance, although
liking other people may relate highly to sociability, it may
also relate to tendencies to be assertive, warm, polite, and responsible. First aggregating such items into a measure of a
structural factor renders such associations almost invisible.
Analysis Strategy
The models outlined in Figures 1A and 1B provide very
different explanations of trait covariation that are not easily compared statistically. For instance, it may not be particularly impressive to show that a large number of FIs do
a better job of explaining behavioral trait covariation than
a small number of structural factors, as a large number of
variables will always better predict an outcome (e.g., obtain
a greater R2) relative to a small number of variables when
other conditions are held equal (Cohen, Cohen, West, & Aiken, 2003). Nor would fixing the number of FIs to the number of structural factors offer a fair test, as structural factors
are identified precisely by their capability to maximally “explain” trait covariation and are not measured independently
of these traits, whereas the FIs are measured with unreliability and independently from the behavioral traits they are associated with.
The overarching goal of these analyses can be considered
to be the identification of a more empirically informed specification of Figure 1B, by identifying specific FIs that are associated with the variation and covariation of different Big
Five-related traits. As noted above, we generally expect that
(a) highly correlated traits will have several distinct FIs that
are associated with both traits; and that (b) uncorrelated
traits will typically also have various FIs that are associated
with both traits, but in a configuration where some FIs relate positively or negatively to both traits (serving to increase
their positive correlation), while others relate positively to
one trait but negatively to the other (serving to increase their
negative correlation).
Study 1: Relations Between FIs and Big Five-Related
Behavioral Traits
Method
A summarized description of the Method is given here,
which is expanded in the Supplementary Materials (S1).
Generating potential FIs involved in shaping Big Fiverelated behavioral trait levels. A total of 529 Wake Forest
University undergraduates completed an online survey for
credit toward a course research participation requirement
(Mage = 18.7; 59% female).
Participants completed the Big Five Inventory (BFI; John
et al., 2008), the Inventory of Individual Differences in the
Lexicon (IIDL; Wood, Nye, & Saucier, 2010), and items from
the IPIP (Goldberg, 1999) identified as highly associated with
Big Five trait levels; scores for each dimension were standardized and averaged. Five to six individuals with the
in
Psychological Review 122 (2015)
highest and lowest levels of each Big Five trait were interviewed for additional credit, resulting in a total of 52 interviews. Interviewed participants were asked to describe the
extent that they performed behaviors related to the trait they
were selected for, and their reasons for doing so. Additionally, approximately half of the participants (N = 229) were
asked to describe someone they knew who was high or low
on one Big Five trait, to describe instances of this person acting in this way, and to suggest their reasons for doing so.
Research assistants extracted reasons for trait-related behavior from the participant interviews and reports of others’
behavior. We considered appropriate reasons for trait-related
behavior to be those that could be classified into roughly the
FI classes described earlier: statements of abilities and efficacies; expectancies of behavioral effects; and of valuations
and goals. A total of 1,985 reasons for performing trait-related behaviors were initially extracted from these sources,
which was reduced to a smaller set of 463 by eliminating
redundancies.
Linking functionality indicators to Big Five-related
traits. A second group of students (N = 511; age M = 18.7;
57% female) completed items describing their levels of these
FIs and of Big Five-related traits via an online survey.
Measures of Big Five-related behavioral traits. Using items
from the BFI (John et al., 2008) and IIDL (Wood, Nye, & Saucier, 2010), we estimated two behavioral traits within each Big
Five trait domain. We excluded items that concerned selfperceptions of valuations or goals, abilities, or expectancies.
We also attempted to measure traits close to the two major
subfacets within each Big Five domain as described by DeYoung, Quilty, and Peterson (2007) and Soto and John (2009).
Following these considerations, the items used to construct
these 10 scales are given in Appendix A; alpha values are
provided in Table 1.
Functionality indicators of Big Five-related behaviors. The 463
reasons identified above were adapted into questionnaire
statements using four different question-response formats.
Items pertaining to likes and dislikes were rated under the
instruction “How much do you like or dislike the following
things?” (1 = strongly dislike this to 5 = strongly like this), and
to goals under the instruction “How much do you try or want
to do the following behaviors?” (1 = I try very hard to avoid doing this to 5 = I try very hard to do this). Items pertaining to
abilities were rated under the instruction “How easy or hard
do you find doing the following things when you try to (or
feel that you should)?” (1 = I find it very difficult to do this to 5
= I find it very easy to do this). All remaining items were rated
under the general instruction “How much do you agree with
each statement?” (1 = strongly disagree to 5 = strongly agree).
The complete item set is available from the first author.
We then reduced the set of 463 items to a more manageable set of approximately 100 items using cluster analysis
techniques (see Wood, Nye, & Saucier, 2010) resulting in the
initial extraction of 87 clusters. We then correlated all 463
items with the 61 items of the IIDL to identify one item from
each initial cluster that best represented the pattern of associations linking the cluster to the IIDL, and to identify additional items that showed particularly large correlations with
a single IIDL item, or correlated above an absolute magnitude of .20 with 10 or more IIDL items, resulting in the identification of 12 additional items for a total of 99 FI items.
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation Results and Discussion
Interrelationships between self-perceptions of Big Fiverelated traits. The interrelationships between the 10 Big Fiverelated traits are shown in Table 1. Generally there were
strong relationships between aspects usually regarded as
being within the same Big Five domain. However, we also
found regular correlations between traits considered to be
in different domains. Consistent with past research (DeYoung, 2006), traits across the agreeableness, conscientiousness, and emotional stability domains tended to correlate
with one another, as did traits across the extraversion and
intellect/openness domains.
Relations between functionality indicators and trait covariation. We explored three main questions concerning the
interrelations of FIs and behavioral traits: (a) which specific
FIs may underlie the covariation found between traits within
the same Big Five trait domain (e.g., sociability and assertiveness); (b) which specific FIs are associated with correlated traits in different trait domains (e.g., sociability and politeness or creativity); and (c) are there specific FIs that are
commonly associated with negligibly correlated behavioral
traits (e.g., assertiveness and politeness)? To explore these
questions, we correlated all 99 FI items with the 10 Big Fiverelated traits. In Table 2, FI items are ordered by the trait
they have the largest association with, except for items #89–
#99, which are placed at the bottom of the table due to having no associations above | r| = .20. To draw out more general themes, we summarize themes of how FI items related
to these traits by paraphrasing the items while indicating
where the full items can be found in Table 2 in parentheses.
Question 1: What FIs are associated with behavioral traits
within the same Big Five domain? For each pair of Big Fiverelated traits, we list the number of FI items that correlate
above | r| = .20 for both traits, and then enter these traits
into a cluster analysis to organize similarities. We then describe each large group of items from these cluster analyses in the text, and also single items that were associated
with both traits but that appeared unassociated with these
clusters. A virtue of this method is that different groups described in the text will generally contribute to covariation independently of one another if entered into a simultaneous
regression, and thus provide relatively distinct paths to the
91
covariation of the two traits.
Assertiveness and sociability. As shown in Table 2, assertiveness and sociability were commonly associated with 21 of the
99 FI items which could be roughly organized in four groups.
The first (#1, #7, #9, #10, #18) largely concerned positive social expectations (e.g., feeling unafraid of social situations).
The second (#6, #8, #11, #12, #16, #14) concerned feeling it
was important and enjoyable to be part of social interactions.
The third (#4, #15) concerned liking and being able to open
up to others. The fourth (#82, #83, #84) concerned ease of
taking risks and trying new things. Fairly distinct from these
groups, individuals who reported being able to put things in
perspective (#74) and follow through with plans (#59), and
liking confronting others (#2) and debating ideas with others (#3) tended to report greater sociability and assertiveness.
Kindness and warmth. Both traits were commonly associated with 35 FI items, in roughly three clusters. The first
(#9, #7, #13, #16, #19, #20, #24) concerned feeling able to be
friendly and comfortable toward others. The second (#21,
#22, #23, #25, #33, #35, #37, #58, #60) concerned perceptions
that one should accept and act dependably toward others.
The third (#29, #38, #39, #40, #41, #42, #43, #44) concerned
finding disrespecting others acceptable/enjoyable. Additionally, individuals who reported trying to see the beauty in
things (#86), and being able to regulate their actions (#64)
and anger (#77), and to act “normally” (#88) tended be report both more kindness and warmth.
Orderliness and reliability. Both traits were commonly associated with seven items, in approximately two clusters. The
first (#52, #55, #59) concerned ease of managing time and
tasks. The second (#50, #51, #57, #65) concerned feeling that
orderly was important.
Anxiety and irritability. Both traits were commonly associated with 12 items, in roughly two clusters. The first (#72,
#73, #74, #75, #77) concerned ease of putting things in perspective and brushing off negative judgments. The second
(#24, #29, #69) concerned perceptions that others were not
helpful or good-natured. Somewhat distinctly, individuals
who reported being unable to keep attention on tasks (#55),
feeling upset when plans change (#68), and who enjoyed
gossiping about others (#42) reported both greater anxiety
and irritability.
Table 1. Relationships Between Big Five-Related Traits.
Trait
Assert
Social
Kind
Polite
Orderly
Relia
Anxious
Irrit
Creat
Unconv
Assertive(.83)
Sociable
.51
(.89)
Kind.11
.44
(.77)
Polite–.09
.24.63(.72)
Orderly
.05.05.07.07 (.86)
Reliable
.11.19.40
.41 .32
(.75)
Anxious
–.14–.16–.21–.26 .04–.08 (.80)
Irritable
.10
–.11
–.36–.48–.04–.16 .56
(.77)
Creative
.28.29.20.14 –.07
.10–.21–.05(.83)
Unconventional
.15–.01–.23–.26–.16–.19–.05 .15 .15 (.36)
N = 511. All |rs| ≥ .09 are significant (p < .05). Correlations above .20 have additionally been shown in bold. Reliabilities are
shown on the diagonal in parentheses. Items used to construct the trait scales are given in the text. “Assert” = Assertive, “Social” =
Sociable, “Polite” = Polite, “Relia” = Reliable, “Irrit” = Irritable, “Creat” = Creative, “Unconv” = Unconventional.
#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
Functionality indicator
L: Being a leader
L: Confronting others
L: Debating ideas with other people
L: Expressing my emotions
G: Correct other people to help them
A: Being talkative in group situations
A: Making friends
L: Being talkative in social situations
P: I feel comfortable around other people
P: I am comfortable around people I don’t know
G: Start conversations with other
L: Surrounding myself with other people
L: Making new friends
L: Being the center of attention
A: Opening up to other people
P: It’s important to take part in conversations in social situations
L: Making “small talk”
P: I am scared to speak up in social situations
A: Being friendly to other people
A: Sympathizing with other people’s feelings
P: Having a positive outlook on life is better for everyone
L: Setting a good example for others
P: It’s important to make other people feel accepted and appreciated
P: People are generally good
P: Having a support system is important
G: Have confidence in myself
A: Accepting people who are different from me
G: Be liked by others
P: Nobody is really there to help me out
P: Talkative people are annoying
L: Silence in social situations
P: I need concrete proof to believe in something
G: Be respectful of others
P: I would feel bad if I insulted someone
P: My friendships are important
P: It’s important to be exposed to a lot of different ideas
P: All people deserve respect
L: Disappointing or angering other people
P: It’s ok to deal with problems by being rude to others
P: There is no good reason to act respectful to others in social situations
P: Friendly people are hard to approach
L: Gossiping about people
G: Come across as being an “angry” person
A: Thinking of good insults
P: I am more important than other people
L: Being stressed out
P: I don’t really care about things that I can’t understand
P: Being worried or stressed would gain me the attention of others
P: People who are disorganized and messy come across as rude
A: Organizing things
P: Organization is necessary
A: Managing my time
L: Making detailed plans
P: Having routines makes me feel comfortable and secure
A: Keeping my attention on tasks
Social
Kind
Polite
Orderly
Relia
Anxious
Irrit
Creat
Unconv
.45.41.22.09 .08 .17–.15–.04 .25
.03
.37.21–.08–.24
–.02–.07–.08.12.08.14
.33.20–.05–.05–.06–.02–.07 .09 .26.25
.29.29.26.01.01–.02.01.16.20
.00
.21
.05–.03
–.07.08.06–.05.09.09.01
.46.62.28.08 .01 .10–.12–.02 .25
.01
.27.56.38.23.00 .17–.24
–.18
.23
–.02
.37.56.25.06–.05 .06–.06 .04 .22
.07
.35.50.38.22–.03 .19–.28
–.18 .10–.01
.34.47.19 .14–.08 .11–.26
–.12
.27
.15
.31.46.29.18 .02 .12–.04–.02 .23
–.02
.29.45.27.10–.03.07.02.02.10–.04
.19
.42.35.23.00 .19–.09–.13 .14–.08
.32.36.08–.07–.06–.13 .02 .14 .23
.10
.22.36.29.09.05.02–.06.01.14
–.08
.22.33.27.20.08 .15–.07–.12 .03–.13
.20.22.08.05
–.04
–.04–.05.02.14.02
–.45
–.50
–.31
–.18–.02–.17 .27
.09–.26
–.03
.03
.33.43.39–.03 .19–.13–.23
.09–.11
–.01
.25.37.28–.03.18–.08
–.10.09.02
.14
.28.36.30.07 .19–.18–.22
.09–.13
.09.16.35.30.08
.29
–.02–.11 .06–.23
–.04.17.33.29–.02.17.01–.06.06
–.09
.01.18.30.26–.04 .08–.20
–.26
.07–.09
.01.14.29.24.09
.24
.02–.07–.01–.11
.17
.20.25.24.06.17–.19
–.15.14.02
.09.16.24.23–.11.15–.14
–.15.17.08
–.01.15.20
.16.02.08.10.00–.05–.20
–.02–.15–.34
–.31
–.06–.16 .20.26–.02.13
–.10–.27
–.28
–.17.06–.12.02.11
–.07.08
–.16–.21
–.22
–.12–.06–.10–.01 .05 .03 .02
.07 .01–.20
–.16.05–.08.01.06
–.05.07
–.03.07.31.37.03
.22
–.09–.12 .05–.06
–.12.06.30.34–.02 .06–.01–.17 .03–.22
.01.18.28.29.05
.25
–.09–.12 .08–.03
.09.14.18.22
–.04.14–.07
–.10.11.12
.00.15.21.21.05 .09–.05–.12 .06–.01
.12–.05–.31
–.39
–.04–.26
–.01.14.00.13
.12–.04–.29
–.39
–.06–.25
.02.18–.03.14
.07–.06–.28
–.32
–.09–.32
–.03.10–.02.10
.03–.12–.29
–.31
–.02–.20
.14
.23
–.02.07
–.03–.07–.22
–.31
–.06–.22.22.28–.11.09
.09–.06–.26
–.30
–.04–.26
.03.19.01.14
.19 .00–.26
–.29
–.04–.08 .01 .21
.11
.22
.16 .03–.19–.28
–.01–.20
–.04.15.02.09
.05–.06–.23
–.25
.02–.18.05.08.01.05
–.01–.03–.19–.23
.01–.11.07.11
–.15.01
–.04–.05–.17–.22
.01–.20
.13 .14–.02–.03
.01–.06–.18–.21
.18–.09.11.14.01.05
.13.18.12.14.60.29–.02–.09 .04–.14
.06.08.07.08.51.23.19 .07–.08–.15
.15.13.04.03.41.28–.09–.07–.04–.14
.12.10.03
–.04.35
.14
.25
.13–.05–.21
–.05 .00–.09–.05 .34
.17
.25
.12–.26
–.26
.08.10.08.11.28.26–.23
–.22
.12–.13
Assert
Table 2. Relationships Between Big Five-Related Traits and FI/Process Items
92 Wood, Hensler Gardner, & Harms
in
Psychological Review 122 (2015)
Relia
Anxious
Irrit
Creat
Unconv
and
P r o c e s s A pp r o a c h e s
to
N = 505. All |rs| > than .09 are significant (p < .05). All correlations greater than .20 are additionally shown in bold. Among functionality indicator items, “L” = Liking item, “G” = Goal item, “A” = ability item, “P”
= perception items, “Assert” = Assertive, “Social” = Sociable, “Polite” = Polite, “Relia” = Reliable, “Irrit” = Irritable, “Creat” = Creative, “Unconv” = Unconventional. Other words in the complete FI items have been
abbreviated for space considerations.
.01–.02–.10–.10–.02–.10 .12 .08–.16–.14
.00–.04 .00 .08–.01 .16–.16–.10 .01–.03
.02.06.02.05.18.10.02–.03–.02–.13
.05–.02–.07–.08–.05–.18–.11–.02 .08 .13
–.04–.03–.10–.09–.15–.19 .09 .05–.08 .04
–.04.00.06.08.04.06.09.04.06–.16
.17.18.14.12–.08.09–.10–.12.10.01
.03.03.11.18.11.14–.14–.17.03–.04
.15.15.08.09.00.10.00–.01.05.02
.10.01
–.08–.13–.06–.03.02.09.05.04
Orderly
90
91
92
93
94
95
96
97
98
99
Polite
.04.09.02.02.25
.16
.24
.15 .02–.07
.05.07.13.14.24.21.10–.05–.06–.16
.02.08.27.27.15
.41
.01–.08–.01–.18
.21.21.21.18
.30.36–.07–.10 .06–.23
.14.18.28.26.10
.33
–.04–.06 .11–.06
.13.18.12.15.15.32
.15–.01 .04–.05
.15.15.10.15.18.31
–.01–.08 .03–.12
.12.12.17.22
.08
.30
–.18–.21
.11–.06
.08.19.21.24.12
.27
–.13–.20
.09–.13
.09.05.11.11.24.27.12.05.01–.19
.11.11.10.15.13.24
.02.01.00–.08
–.02–.05–.19–.25
–.12–.27
.11.16–.16.03
–.01–.09–.23
–.22.21–.02
.48.33–.08–.10
–.15–.27
–.30
–.28
–.08–.22.40.31–.17.04
–.05–.02
–.02.01.13.11.31
.18–.16–.10
.02–.05–.03–.08 .09 .00 .21
.19–.10 .03
.18.08.03.10–.05.02–.57
–.35.20.05
.24
.14.04.05
–.05.04
–.36
–.21
.15.18
.20.22.18 .19–.03 .15–.36
–.29.20.02
.15
.22
.19 .15–.04–.01–.35
–.30.22–.05
.08.11.17.16.04.11–.20
–.13.10.05
–.10.03.23.30.04 .08–.30
–.43
.05–.09
.15
.21
.14 .11–.03 .02–.18–.05 .73
.14
.04 .10 .07 .02–.01–.07–.18–.01 .52
.14
.12 .09 .05 .11–.13–.07–.18–.04 .48.30
.02 .08 .05 .09–.02–.01–.15–.05 .41
.16
.34.28.15 .05–.10 .05–.14–.01 .39.24
.28.36.21.18–.07 .06–.29
–.20.37.16
.26.24.01–.08–.06–.12–.25
–.01
.31.25
.16
.28.26.22–.11 .08–.20
–.12
.30
.11
.04.15.26.26–.03 .08–.11–.12 .27
.02
.21
.16–.05–.12–.12–.09–.03 .10 .19 .26
.02.11.24.26.13
.23
–.15–.19–.12–.29
.11 .14 .01–.02–.12–.01–.09–.02 .16 .15
Kind
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
Social
L: Being a perfectionist P: Time management is important
P: It’s important to be reliable and dependable
A: Following through with plans
P: Other people depend on me
P: I pressure myself to be as successful as possible
L: Working hard
A: Thinking rationally
A: Regulating my actions
G: Be “on top of things” at all times
G: Achieve my career goals
P: I don’t feel the need to learn new things
P: I get stressed out if my plans change
P: I feel that I won’t be accepted by others
P: I worry about messing up on school assignments
L: Having other people know when I’m right
A: Staying calm in potentially stressful situations A: Brushing off what other people think of me
A: Putting things into perspective A: Seeing the positive side of bad situations P: Staying calm in situations is beneficial for everyone
A: Managing my anger
A: Coming up with creative ideas
A: Creating visual art (drawing, painting)
L: Abstract ideas A: Appreciating art
L: Standing out by being different from the crowd
A: Trying new things
A: Taking substantial risks
G: Try new things
G: See beauty in everything
L: Doing strange things to get a reaction out of others
A: Acting normal
L: Making jokes
P: It’s better to find a single right answer to an issue than to consider
multiple interps.
A: Understanding math and science
A: Getting work done in silence
P: Getting good or bad grades is not going to affect what I want to do in life
P: I need pressure to get anything done
P: I get uncomfortable when others tell offensive jokes
P: If I help others, they will also help me
G: Maintain self-control in stressful situations
P: Others respond well to people who are outgoing
P: Nobody really pressures me to work hard
Assert
#
Functionality indicator
Table 2 (continued)
Functionalist
Behavior Explain Trait Covariation 93
94 Wood, Hensler Gardner, & Harms
Creativity and unconventionality. Both traits were commonly associated with 5 items. The most highly associated
items (#3, #80, #82) concerned enjoying abstract ideas, debating ideas with others, and standing out from others. Somewhat distinctly, individuals reporting greater ease in taking
risks (#84) and who disagreed that routines helped make
them feel secure (#54) tended to report more creativity and
unconventionality.
Question 2: What FIs are commonly associated with correlated
behavioral traits in different Big Five domains? We next conducted
analyses to explore the FIs that might underlie the covariation
of traits typically categorized under different Big Five factors.
We first explored FIs associated with both sociability and politeness (r = .24), and sociability and creativity (r = .29), again
focusing primarily on items associated above |r| ≥ .20 with
both traits. We then explore FIs that were correlated with traits
in several Big Five domains simultaneously.
Sociability and politeness. Both traits were commonly associated with ease of being friendly (#7, #19, #20); expecting
comfort and acceptance from others (#9, #69); valuing a positive outlook (#21) and socializing (#16); and greater goals
of trying new things (#85), having self-confidence (#26), and
making friends (#13).
Sociability and creativity. Both traits were associated with
increased enjoyment, effort, and ease of engaging and expressing oneself in social interactions (#1, #3, #4, #6, #7, #8,
#11, #10, #14, #18, #82); with ease of putting things in perspective (#74, #75), and of coming up with creative ideas
(#78); and with both the desire and ease of trying new things
and taking risks (#83, #84, #85).
Question 2B: What FIs are associated with behavioral traits
in several Big Five domains simultaneously? Paralleling other
studies that have examined the intercorrelations among behavioral traits, we can see in Table 1 that there were almost
invariably significant positive correlations between every
pairwise combination of sociability, kindness, politeness, reliability, creativity, and assertiveness, and negative correlations between these traits with anxiety, irritability, and unconventionality. Although this is sometimes interpreted as
an artifact of the blending of orthogonal factors (Ashton et
al., 2009) or the product of higher order factors (DeYoung,
2006), this phenomenon can alternatively be understood as
indicating the existence of FIs which influence the functionality of many distinct behavioral traits simultaneously. As
can be seen from Table 2, numerous FIs were associated with
most traits simultaneously. We list some themes below.
Ease in social situations. Items concerning ease and comfort
in social situations (#6, #7, #9, #10, #18) tended to be most
associated with sociability and assertiveness, but also related
positively to kindness, politeness, and creativity, and negatively to irritability and anxiety.
Negative expectations of social situations. Individuals who
expected that others would not accept or help them (#29,
#69) especially reported lower kindness and politeness and
greater anxiety and irritability, but also reported somewhat
lower assertiveness, sociability, reliability, and creativity.
Importance placed on acting positively. Individuals who felt
that others depended on them (#60), that they should have a
positive outlook (#21) and take part in social situations (#16),
and who tried to be successful (#61) and develop self-confidence (#26), tended to be more sociable, kind, polite, assertive,
reliable, and creative, and less anxious and irritable.
in
Psychological Review 122 (2015)
Interpersonal skills. Individuals who reported greater ease
in acting “normal” (#88) and friendly (#19), accepting others (#27), and seeing the positive side of situations (#75), reported greater kindness, politeness, sociability, and reliability, and less anxiety and irritability.
Self-regulatory and task related abilities. Individuals who reported greater ease of regulating their actions (#50, #55, #64),
and who liked working hard (#62) tended to report being
more reliable, orderly, kind, polite, and sociable, and less
anxious, irritable, and unconventional.
Importance and ease of seeing and doing things differently. Individuals who placed greater importance in seeing different perspectives (#36) and trying new things (#85), and who found
it easy to come up with creative ideas (#78), tended to report
especially higher creativity and unconventionality, but also
higher assertiveness, sociability, kindness, and politeness.
Enjoyment of being different. Individuals who enjoyed being different from the crowd (#82) and provoking reactions
from others (#87) particularly reported being more creative
and unconventional, but also being more assertive, sociable,
and less orderly.
Question 3: What FIs are commonly associated with uncorrelated behavioral traits? As we have suggested, even uncorrelated traits should regularly share FIs in an arrangement
such that some FIs show same-sign correlations and others
opposite-sign correlations in about equal magnitude. We explore this idea by examining two trait pairs that show nearzero correlations in Table 2: orderliness and anxiety (r = .04),
and assertiveness and politeness (r = −.09). In the text, we
discuss FIs that were significantly associated (p < .05) with
both traits simultaneously.
Orderliness and anxiety. Tendencies to be orderly and
anxious were commonly associated with 23 FI items. Both
were positively associated with believing order was important (#49, #51, #57), and with concern about messing up on
tasks (#70) or changing plans (#68). Orderliness and anxiety
were both higher among people who liked structure (#53,
#54, #65), being a “perfectionist” (#56), and being seen as
“right” (#71), and were both lower among people who liked
abstract ideas (#80), being different from others (#82), and
who strove to try new things (#85). However, ease of managing time (#52), staying on task (#55), self-regulating (#64),
and acting “normally” (#88) were associated with increased
orderliness but decreased anxiety.
These results thus indicate that concerns and preferences
for order may serve to increase both orderliness and anxiety, but abilities to create or maintain order may serve to increase orderliness but decrease anxiety. These FIs appear to
suppress one another and result in the near-zero association
between these traits.
Assertiveness and politeness. Tendencies to be assertive and
polite were commonly associated with 39 FIs. Both were positively associated with ease and enjoyment of socializing (#7,
#9, #10, #11, #12, #13, #16, #18, #69); effort and enjoyment
in doing well (#61, #62, #65, #66); and ease of thinking creatively and rationally (#63, #78), putting things in perspective (#74, #75), staying organized (#50) and staying calm in
tense situations (#72). Both were negatively associated with
perceiving talkative people as annoying (#30) and enjoying silence (#31). However, perceiving more ease and value
in confronting or provoking others (#2, #34, #38, #39, #43,
#44, #45, #87) was associated with greater assertiveness but
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation 95
conscientiousness, fretful and temperamental in the domain
of neuroticism, and creative and philosophical in the domain
of openness. Self-rating estimates of these traits were formed
by aggregating the multiple self-ratings of these items over
the course of multiple years of the study; peer rating scales
were formed by aggregating the 1 to 3 peer ratings.
International Personality Item Pool (IPIP). ESCS participants
completed up to 2,492 IPIP items between Spring, 1994 and
Spring, 2003, which consisted of short items describing their
behavioral traits, feelings, skills, beliefs, and more abstract
self-perceptions.
The first and third authors and a research assistant categorized the IPIP items as either an FI item (e.g., abilities, beliefs, likes/dislikes, goals, values) or a behavioral trait item. For
the 1,351 items classified as FIs, the item’s correlation with
a self-reported and peer-reported SMM trait was first averaged. We then identified the 100 FI items with the highest absolute correlation with each SMM trait. Across the 10 traits,
this resulted in a smaller set of 630 items. These items were
then entered into a cluster analysis, which was prespecified
to extract 75 clusters. These clusters ranged in size from 24
items, to a one-item cluster which was not considered further. For the remaining 74 clusters, we selected a single item
to represent in Table 4 which was most representative of how
the cluster was associated with the 10 traits.
decreased politeness, and ease of managing anger (#77) was
associated with higher politeness but lower assertiveness.
Across both examples, we see that behavioral traits regularly share many common FIs even when they are almost
perfectly uncorrelated, arrayed such that some showed
same-sign associations and others opposite-sign associations with the two traits. This is consistent with our understanding that uncorrelated behavioral traits should generally
be expected to be influenced by common FIs that suppress
one another, rather than by entirely separate systems of FIs.
Study 2: Relations Between FIs and Big Five-Related
Traits Within the IPIP
The IPIP (Goldberg et al., 2006) consists of a set of nearly
2,500 items completed by participants in the ESCS. Most participants have both self- and peer-ratings of Big Five-related
traits similar to those examined in Study 1, affording the potential to test whether the major contours of relations between
FIs and Big Five-related traits described in Study 1 replicate
in a new sample, and across both self- and other-perceptions.
Again, we present a summarized description of the Method
which is expanded in the Supplementary Materials (S1).
Method
Participants. A total of 700 ESCS participants completed
the materials examined here as part of an ongoing study.
Participants ranged in age from 18 to 85 (M = 51.4, 56% female), and were of all levels of education (see Goldberg, 2008
for additional details.
Materials: Saucier Mini-Markers. In this study, we utilized
the Saucier Mini-Markers (SMM; Saucier, 1994) as our measure of personality as it was administered multiple times in
the form of both self-ratings and peer-ratings over the course
of the multiyear study. There were up to three self-ratings
and up to three peer-ratings of the SMM that could be used
for each participant.
We used indicators of two distinct traits from each Big
Five domain selected to parallel the traits examined in
Study 1 to the extent possible, and which were examined
separately: the items bold and extraverted within the domain of extraversion; kind and cooperative in the domain
of agreeableness; organized and practical in the domain of
Results and Discussion
Question 1: What FIs are associated with behavioral
traits in the same Big Five domain? We first examined FIs
associated with traits regarded as being within the same trait
domain. In the text, we only discuss items that were both (a)
correlated at least | r| = .20 with either self-ratings or peerratings of both traits, and (b) significantly correlated with
both self- and peer-ratings of these traits. We entered these
items into a cluster analysis to organize redundancies, and
in the text identify the resulting groups by their item numbers in Table 4.
Bold and extraverted. We found 11 items associated with
both traits, in roughly three clusters. The first (#1, #6, #10,
#13, #16, #18) consisted largely of reported ease of influencing others, expressing feelings, and approaching others. The
second (#2, #12) consisted of enjoyment of power and attention. The third (#7, #48) consisted of reported ease of making
Table 3. Relations Between Self- and Peer-Rated Big Five-Related Trait Perceptions
# Trait
αS
αP
1
2
3
4
5
6
7
8
9
10
1 bold
2 extraverted
3 kind
4 cooperative
5 organized
6 practical
7 fretful
8 temperamental
9 creative
10 philosophical
.76
.43
.54
.49
–.06–.10.07–.05–.14.15.14.07
.85
.53
.45
.54
.09.08.06–.03–.10.00.14.07
.66.27.04.12.23
.47
.09
.21
–.09–.36
.07.10
.62.35
–.02.12.52
.20
.22.30
–.21
–.45
.09.17
.86.57.09.07.10.16.50
.38
–.13–.09.05.04
.72.45.04–.04.17.34.38.30
–.23
–.26
–.07.09
.72 .43–.08–.06–.09–.12–.13–.13 .30
.35
–.04–.11
.77 .48 .07–.01–.22
–.29
–.15–.19 .44
.37
.05–.16
.85
.52
.29.20.15.12.05.08–.11–.07.44
.25
.81.49.10.09.12.09
–.01.05–.02–.07.31
.42
N ≥ 590. αS = Alpha from self-ratings, αP = Estimated alpha from peer-ratings. All |rs| ≥ .08 are statistically significant (p < .05),
all |rs| ≥ .20 are additionally shown in bold. Correlations between self-ratings are shown below the diagonal, and correlations
between peer-ratings are shown above the diagonal. Correlations between the self-rating and peer-rating for the same item are
shown on the diagonal and italicized.
#
Have a natural talent for influencing people
Want to be in charge
Love a good fight
Believe that I am important
Can handle complex problems
Am good at making impromptu speeches
Make decisions easily
Let myself be pushed around
Try to please everyone
Usually like to talk a lot
Love large parties
Like to attract attention
Would make a good actor
Usually enjoy being with people
Am considered attractive by others
Have difficulty expressing my feelings
Have difficulty showing affection
Find it difficult to approach others
Like to help others
Love to make other people happy
Love children
Feel badly if my . . . actions cause
someone . . . to feel emotional pain
Am good at understanding others’ feelings
Am concerned about others
Believe the poor deserve our sympathy
Love dangerous situations
Don’t care about dressing nicely
Don’t care about the rules
Feel that life has no meaning
Pay attention to details
Believe that planning ahead makes things
turn out better
Can make myself work on a diff. task even
when I don’t feel like it
Want everything to be ’just right’
Find it difficult to organize tasks and
activities
Have difficulty starting tasks
Am easily distracted Am not highly motivated to succeed
Believe in a logical answer for everything
Can’t hold onto money for long
Can’t afford to buy things I need
Feel a sense of worthlessness or
hopelessness
Dislike myself
Worry about my health
Often have the feeling that others laugh or
talk about me
S
P
S
P
S
P
S
P
.13
.25
.11–.10–.04–.11–.08
in
.30
Wood, Hensler Gardner, & Harms
.13
.11–.05–.03–.07–.07 .00–.04 .09 .03
–.09–.04–.08–.12–.15 .01–.16–.04–.17 .00–.19 .00 .33
P
44
S
–.15–.08–.16–.17–.14–.05–.16–.10–.20
–.12–.19–.20.33.27.32.17–.12–.02–.08–.02
–.20
–.12–.18–.11–.19 .02–.15–.07–.23
–.12–.21
–.14
.37.20.28.16–.12–.01–.07 .01
–.03–.05–.03–.03–.08–.04–.09–.05–.06–.07–.10–.08 .30.22.27.15 .02 .01–.01–.11
Philos.
41
42
43
P
–.14–.12–.10–.13 .01 .08–.08–.02–.45
–.30
–.22
–.11.16.11.13.04
–.10–.07–.07.03
–.15–.11–.15–.05–.07 .07–.10 .05–.31
–.15–.18
–.06.17.12.12.01
–.18–.07.07.02
–.06–.06–.02 .05–.04 .07–.05 .04–.28
–.14–.13–.06 .21
.09.18.04–.06.02
–.02.01
–.19–.14–.23
–.10–.06 .01–.09 .02–.25
–.13–.18–.08 .10 .09 .09 .06–.14–.04–.11–.01
.02–.02–.11–.07–.07–.06 .07–.02 .17 .13 .23
.16–.02–.07–.02 .01–.05–.16–.10–.16
.01–.02 .02–.02–.05 .04–.11–.03–.25
–.15–.31
–.26
.13.10.20
.09.06.12.00.00
–.11 .00–.07–.05–.03–.03–.10–.01–.14–.09–.17–.20
.16.03.13.12.03.05
–.04–.06
S
Creative
34
35
36
37
38
39
40
P
Temper.
.14.04.02–.06.06
–.02.14.01.19.10.21
.09–.17–.06–.15–.08 .19 .04 .10–.04
–.01.03
–.02.10.08.04.04.04.32.22.15.07.13.03.10.05–.02–.04–.07–.03
S
Fretful
32
33
P
Practical
.07.05.04–.03.09
–.02.14.11.30
S
Organized
31
P
Cooper.
–.09–.06 .01–.05 .27
.15
.21
.08.01
–.01.04
–.01–.07.06
–.15–.10.06.08.11.04
.10.11.21
.18
.31
.10.15.02.03–.10.03
–.17–.05.05
–.10.00.15.19.16.04
–.05.01.04.08.36
.19
.23
.13.02
–.01.07
–.04–.05.04
–.07–.06.08.05.09.07
–.06
–.01.00.01.19.16.15.10
–.04–.04.03.04–.04
–.01–.05–.07
–.04.05.19.11
.22
.06 .04 .01–.17–.21
–.19–.14–.07–.13–.06–.07–.02–.11 .09 .06 .00–.06–.01–.02
.07 .04–.14–.07–.22
–.08–.22
–.10–.13–.11–.06 .03 .00–.11 .10 .05–.02–.03–.05–.02
.13 .12–.02 .07–.15–.14–.21
–.11–.17
–.11–.12
–.13.04.01.08.05.10.10.12.10
–.10–.11–.15–.13–.16–.13–.18–.13–.10–.06–.07–.08 .08–.01 .14 .09–.11–.01–.05 .01
.09.07.05.01.10
–.02.18
–.04.37
.18
.32
.10–.08 .00–.14–.01 .21
.05.10.06
S
Kind
22
23
24
25
26
27
28
29
30
P
Extrav.
.44.31.40.31.08–.06 .04–.05 .10–.01 .01–.07–.07–.05 .03 .05 .30
.10
.23
.16
.38.31.31.26
–.05–.10
–.03–.09.16.10–.01–.03–.02–.11.05.06.18.06.09.06
.31.21.15 .04–.13–.18–.19–.14–.04 .02–.07–.08 .03–.08 .19 .12 .04–.05 .08 .04
.29.22.24.11.13
–.01.11.02.14.05.15–.01
–.14–.14–.08–.04.17.06.08.02
.31.20.13.04.07
–.12.07
–.02.19.06.16
–.01–.26
–.19–.14–.08 .31
.10
.22
.15
.37.32.38.26.01–.04 .00 .01 .08–.04 .01–.03–.14–.12–.07–.03 .25
.09
.22
.16
.35.25.23.19–.01–.09 .03–.06 .16 .09 .08–.03–.22
–.17–.11–.02 .20
.09.08.01
–.23
–.19–.12–.06–.05 .09 .04 .08–.24
–.17–.14–.02 .21
.10 .08–.08–.17–.05–.02–.01
–.16–.20
–.04.01.14.16.18.16
–.01.02.02.02.15.02.04
–.06–.12
–.05–.08
–.09
.26.30.48.37.06.00.02–.08
–.01–.04–.03–.11.08.12.10.17.04.01.00.01
.17.16.30.31.02.03.03.00.04.04
–.05–.03–.02–.01.00.04.02.01
–.01–.03
.31
.19
.35.21
–.08–.09–.07–.08–.04–.03–.07–.14 .05–.02 .13 .04 .13 .02 .11 .06
.26
.17
.32.20.02–.04 .03–.04–.01–.05–.04–.13 .04 .00 .06 .03 .20
.06
.20
.11
.09.17.34.28.25.11
.22
.09 .10 .02 .08 .01–.14–.09–.14–.12 .03 .04–.01–.06
.22
.13
.26
.19 .14–.04 .13–.02 .12 .03 .10–.03–.09–.03–.03 .02 .21
.12.12.09
–.25
–.26
–.37
–.34
–.14–.04–.11–.01–.17–.09–.01 .10 .14–.04 .08–.02–.19–.15–.20
–.07
–.05–.13–.25
–.19–.28
–.14–.21
–.07–.07 .00–.02 .09 .06–.05 .06 .00–.09–.14–.08–.10
–.27
–.26
–.40
–.27
–.10.00
–.10.01
–.12.02
–.04.07.18.07.12–.01–.20
–.09–.12–.15
–.01.03.15.13.34.25.26.16 .05 .00–.01 .01–.07 .01–.05–.08 .04 .05–.01–.05
–.04.06.14.17.36.20.26.19.05.00.04
–.04–.01.03.00–.04.04.12.05.01
.00
–.02.09.11.29
.17
.22
.15 .04–.01–.01–.02–.07–.05–.07–.13 .09–.04 .05–.03
S
Bold
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
IPIP FI item Table 4. Relations Between Self- and Peer-Ratings of Big Five-Related Traits and FI/Process IPIP Items
96 Psychological Review 122 (2015)
#
S
P
S
P
S
P
.02.03.14.05.09.00.11
–.04–.03–.03–.01
–.13.01.09
–.06.01.12.13.29
.00–.04.00
–.01.06.01.09.06.13.17.17.12.01.03.01.01–.01
–.02–.21
–.15–.07
–.05–.05.02.04.07.04.03.05.03.07.07.07.00
–.02–.22
–.12–.42
–.14–.20
–.07–.09 .04–.02 .02 .07 .05 .05 .05 .09–.16–.05–.16–.08–.13–.18–.29
–.20
–.12–.07–.03.11.10.17.07.14.12.11.10.02.06
–.03–.05–.19–.15
–.24
70
71
72
73
74
.28
.25
.26
.24
.18
–.24
–.28
–.20
–.22
.18
to
.14
.13–.02–.06–.04–.04
.14–.03 .02 .06–.04
.16–.03–.04 .01–.07
P r o c e s s A pp r o a c h e s
.21.21.13 .17 .05 .11–.03 .06–.04 .02–.10–.10–.11–.08–.08 .22
.08
.07
.08
–.11–.10–.18–.11–.07
.18
–.11–.17–.11–.13–.07–.10–.08 .21
–.27
69
.18.19.08–.16.00
–.04.00
.14 .18 .12–.08–.03 .01–.02
.17
.26
.08–.14 .02–.05–.01
.18
.21
.07–.17–.07–.07–.06
–.16–.22
–.10.17.09.05.02
–.18–.24
–.12–.01–.02 .01 .03
–.23
–.18–.09.10
–.01.06.05
P
.01.07.06.03.20
.07 .09–.03 .02–.15 .01–.15–.04 .08–.01 .10 .29.23.25.04
.19 .11 .18 .08 .01–.05–.02–.11–.15–.19–.06–.17–.10–.02 .03 .03 .27.21.16.15
–.15–.12–.10–.03–.09–.02–.02 .00–.04 .04 .01 .10 .15 .08 .07–.01–.37
–.25
–.23
–.15
.12.18.14.07.14.04.09
–.04.00
–.11.05
–.09.01.04.01.02.22
.15
.34
.17
.17 .01 .04 .01–.10–.03–.13–.02–.14–.19–.21
–.16.11
–.05.17.05.18.15.27.20
S
64
65
66
67
68
P
.10.11.14.08.14.05.11
–.03–.08–.17–.03
–.17–.03.02
–.05.00.30.25.35.17
.07.04.01–.02.11
–.03–.03–.02–.03–.08.01
–.01.04.00.04.05.19.22.22.06
S
62
63
Philos.
.05 .02–.07–.05–.14–.01–.16–.09 .17 .14 .05 .05 .22
.09
.24
.18–.02–.03–.04 .01
.08.03.16.11.23
.13
.23
.11.11
–.03.12.02–.24
–.16–.24
–.19.07.03.03–.03
–.05–.05.03.01.19.13.17.16
–.02
–.09–.02.02
–.30
–.15–.41
–.24
.06.02.04.01
–.05.01
–.02–.04.11
–.01.10.09.24
.11.14.16
–.21
–.14–.29
–.15 .05–.03–.03 .00
–.11–.06–.04–.03 .22
.11
.20
.09.04.00.07.06–.15–.05–.23
–.19–.02.02.01.10
.24
.10 .10 .06 .08–.01 .02–.06 .03–.05 .00–.08–.06–.05–.01 .04 .38.25.22.13
P
56
57
58
59
60
61
S
Creative
.06 .04 .01 .04–.14–.06–.21
–.11 .03 .07–.08–.02 .21
.07 .07–.11–.09–.18–.12–.18–.13–.13–.05–.08–.06 .25
P
Temper.
54
55
S
Fretful
.00–.07–.12–.13–.21
P
Practical
53
S
Organized
–.09–.06
–.04–.02
–.04.00.01.06.06.02.07.19
–.16
–.23
P
Cooper.
52
S
Kind
–.17–.07–.18–.13–.15–.06–.17–.15–.13–.07–.13–.13 .27
–.06–.07–.12–.09–.14–.04–.13–.07–.10–.08–.07–.07 .24
–.12–.04–.06 .04 .00 .04–.07–.04–.09–.01–.08–.03 .33
–.23
–.15–.22
–.21
–.08.04
–.05.02
–.07.02
–.04.03.31
.20
.11.19.10.10.01.18.04.07–.01.09
–.05
–.28
.08.04.08.08.08.00.12.08.19.09.17.14
–.21
.23
.17.08.09.07
–.04.11.04.12.09.17.09
–.23
P
Extrav.
45
46
47
48
49
50
51
S
Bold
and
For self-reports, all N ≥ 659, and all |rs| ≥ .08 are significant (p – .05). For peer reports, all N ≥ 472, and all |rs| ≥ .09 are significant (p < .05). All significant correlations are underlined; all correlations greater than |r| =
.20 are shown in bold.
“S” = self-perception, “P” = peer perception, “Extrav.” = extraverted, “Cooper.” = cooperative, “Temper.” = Temperamental, “Philos.” = Philosophical. Other words in the complete Functional IPIP items have been
abbreviated for space considerations.
Remember my failures more easily than
my successes
Expect things to fail
Feel crushed by setbacks
Worry about being embarrassed Adapt easily to new situations
Find it easy to stay healthy
Can stand a great deal of stress
Experience very few emotional highs and
lows
Feel that friendly people are actually trying
to manipulate me
Tend to become agitated when I have to sit
and wait for something
Suspect hidden motives in others
Get upset if others change the way that I
have arranged things
Look on the bright side
Am not easily annoyed
Easily resist temptations
Don’t try to get even
Love to think up new ways of doing things
Am usually aware of the emotions that are
portrayed in various types of art (for ex.,
painting, photography, music, dance)
Enjoy feeling “close to the earth”
See beauty in things that others might not
notice
Prefer variety to routine
Have difficulty imagining things
Try to understand myself
Enjoy wild flights of fantasy
Look forward to the opportunity to learn
and grow
Enjoy discussing movies and books with
others
Believe that we should be tough on crime
Am not interested in theoretical discussions
Don’t try to figure myself out
Would hate to be considered odd or strange
IPIP FI item Table 4 (continued)
Functionalist
Behavior Explain Trait Covariation 97
98 Wood, Hensler Gardner, & Harms
decisions without concern of embarrassment. Somewhat independently, individuals who believed they were important
(#4) and attractive (#15), and who looked forward to opportunities to learn and grow (#69) tended to be both more bold
and extraverted.
Kind and cooperative. We identified five items associated with
both traits. Four (#19, #20, #21, #24) pertained to feeling it
was important and enjoyable to help others, and the latter concerned a tendency to see the positive side of situations (#57).
Organized and practical. We identified five items associated with both traits. Three (#30, #34, #39) pertained to ease
of staying organized and maintain resources. Somewhat distinctly, individuals who disliked themselves (#42) and who
did not think planning was important (#31) tended to be
both less organized and practical.
Fretful and temperamental. We identified 10 items associated
with both traits, in roughly two clusters. The first (#41, #42,
#44, #57, #49) concerned disliking oneself and feeling disliked
by others. The second (#43, #50) pertained to concerns about
health. Somewhat distinctly, both were associated with ease of
being annoyed (#58) and of resisting temptations (#59).
Creative and philosophical. We identified eight items associated with both traits, in roughly two clusters. The first (#1,
#5, #61, #66, #72) largely concerned interest in and ease of
solving complex, theoretical, or novel problems. The second
(#62, #67, #69) largely concerned desires to learn about oneself and to grow and develop.
Question 2: What FIs are associated with traits in diverse trait domains? Similar to Study 1, we next identified
FI items that were correlated at a | r| ≥ .10 level with traits
in several Big Five domains simultaneously.
Negative beliefs about oneself, others, and the future. Tendencies to dislike oneself and feel unimportant (#4, #41, #42,
#45) were most associated with neurotic tendencies, but were
also associated with lower levels of nearly every other trait
examined. Items indicating negative expectations about others (#44, #53, #55) and the future (#29, #46, #57) showed
similar associations.
Self-regulatory skills. Similar to Study 1, individuals who reported greater ease of organizing and completing tasks (#7,
#32, #34, #35) were particularly more organized, but were also
more practical, bold, extraverted, and less fretful, both in selfand peer-reports, and described themselves as more cooperative, creative, and philosophical, and less temperamental.
Interpersonal skills. Similar to Study 1, individuals who
reported skills at expressing and understanding themselves
and others (#16, #17, #18, #23, #49) were more extraverted,
bold, kind, and creative in both self- and peer-reports, and
also described themselves as less fretful and temperamental
and more philosophical.
Valuation of traditions. Similar to Study 1, people who disliked rules and routines (#28, #65) and were unconcerned
with being considered “strange” (#74), were more bold, creative, philosophical, and less kind, cooperative, organized,
or practical, in both self- and peer-reports.
Question 3: What FIs are associated with uncorrelated or
weakly correlated traits? As in Study 1, we were interested
in illustrating how negligibly correlated traits can share common FIs in a suppressive manner. We returned to roughly
the same trait-pairs examined in Study 1, and only discuss
FIs that were significantly associated with either self-ratings
or peer-ratings of both traits.
in
Psychological Review 122 (2015)
Organized and fretful. As in Study 1, there were several FIs
with same-sign correlations with both traits. Tendencies to
want order (#33) and to be upset with disorder (#56) were
associated with being both more orderly and fretful, while
enjoying danger (#26), variety (#65), and being unconcerned
with dressing nicely (#27) were associated with both being
less orderly and less fretful. However, other FIs showed opposite-sign correlations. Individuals who disliked themselves
and were suspicious of others (#41, #42, #44, #45, #55), who
had difficulties handling interpersonal and complex situations (#5, #16, #18), and who found it difficult to start and
stay focused on tasks (#32, #34, #35, #36) tended to be less
organized and more fretful.
Bold and cooperative. As in Study 1, reported abilities to
do difficult or novel tasks (#18, #32, #35, #49, #51) and to
express and understand emotions (#16, #17, #23, #62) were
positively associated with both traits; and feelings that life,
others, or oneself were dislikable or without value (#4, #15,
#29, #41, #42, #45) were negatively associated. However,
enjoying danger (#3, #26) and disliking conformity, rules,
and routine (#9, #28, #65, #68, #74) was associated with increased boldness but decreased cooperativeness.
Summary of the Empirical Illustrations
As noted at the onset, a major goal of the empirical illustrations was to yield a more empirically grounded specification of Figure 1B, that is: of the FIs that may be involved
with shaping levels of many distinct behavioral traits simultaneously. Consequently, we conclude these illustrations by
identifying 22 specific FIs which appeared to both (a) be relatively similar across the two studies, and (b) relate to trait
perceptions in a relatively similar manner in Tables 2 and
4. These are presented in Figure 2. For instance, “ability to
imagine things” was positively associated with assertiveness, sociability, warmth, politeness, and negatively associated with anxiety in both studies (#78 in Table 2, #66 in Table 4). These items were cluster analyzed using the relevant
items from Study 1 in order to place more highly covarying
items closer together in the figure; this suggested these 22 FIs
could be organized in roughly five clusters.
This diagram illustrates some important overarching
themes from the two empirical illustrations. First, traits in a
common trait domain seemed to covary due to the sharing of
numerous distinct FIs. For instance, the present results suggest that assertiveness and sociability covary in part because
both traits are similarly facilitated by FIs such as liking attention, and social situations; by the ease of opening up to others, and of adapting to new situations; and by expectations
that interactions with others will go well. Many of these FIs
are relatively distinct from one another and thus should explain independent aspects of the covariation between assertiveness and sociability. Second, we found that associations
between a particular FI and trait perceptions were rarely localized to a single Big Five domain. Rather, FIs regularly had
associations across multiple trait domains simultaneously.
Third, even those FIs that had higher semantic and statistical
similarity to one another (as indicated here by being on the
same cluster) had distinguishable associations with trait perceptions that were relatively consistent across the two studies. For instance, “liking to attract attention” and “liking being considered odd/strange” might be considered as being
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation 99
Figure 2. The left half of the diagram depicts some of the FIs found to relate to Big Five-related traits in a similar fashion in both
Studies 1 and 2. Subscripts indicate where the corresponding FIs are listed in Tables 2 and 4, respectively. Solid lines indicate positive effects, and dashed lines indicate negative effects. The right-half of the diagram, adapted from Wood (in press), depicts a conception of the Big Five structural factors as “useful summaries” of ways in which an individual’s traits impact relationship decisions regarding the individual made by others (e.g., whether to befriend or begin a romantic relationship with the individual). See
Footnote 1 regarding how to interpret relations between behavioral traits and trait perceptions (e.g., assertiveness, sociability) and
broader structural factors (e.g., extraversion).
fairly similar, but the former had positive associations with
sociability not found with the latter, and the latter had negative associations with orderliness not found with the former.
It is also reassuring to find that the relationships between
particular FIs and Big Five-related traits seemed to generally be similar to relationships identified previously in the
literature. For instance, Connor-Smith and Flachsbart (2007)
reported that the ability to cognitively restructure negative
events was associated with more positive levels of all Big
Five traits, a finding paralleled here (#74, #75, #20 in Table
2; #23, #57 in Table 4). Similarly, Wood, Harms, and Vazire
(2010) found tendencies to see others positively to be associated with more positive levels of traits in all Big Five domains, a finding paralleled here (#24, #29, #96 in Table 2;
#44, #53, #55 in Table 4).
An important caveat of these empirical illustrations is that
the documented relationships between FIs and behavioral
traits have to indicate causal relationships—with FIs causally impacting behavioral trait levels—in order for FIs to explain trait variation and covariation in the manner conceptually described here (Figure 1B). The cross-sectional nature
of the two explorations makes it impossible to definitively
establish the directionality of these relationships. However,
it would be very surprising if the associations documented
in Tables 2 and 4 represented mainly noncausal effects of FIs
on behavioral traits for several reasons.
First, a diverse range of theoretical models posit FIs as
the proximal causes of an individual’s behavior (e.g., Ajzen,
1991; Feather, 1982; Gintis, 2007; Mischel & Shoda, 1995).
These associations appear to be generally consistent with
the overarching functionality assumption that FIs relate to
behavioral tendencies by how they alter the behavior’s association to desired ends. For instance, it is easy to imagine
how a “belief that rudeness is socially acceptable” would decrease the perceived costs of behaving in an assertive, unkind, and irritable manner, and shape levels of these behavioral tendencies accordingly. Second, it is increasingly clear
that cross-sectional relationships between FIs and behavioral traits frequently replicate in within-person studies, or
when the associated FIs are directly manipulated (Dweck &
Leggett, 1988; Fleeson, Malanos, & Achille, 2002; Higgins,
2000; McCabe & Fleeson, 2012; Metcalfe & Mischel, 1999).
Third, behaviors, abilities, expectancies, motives, and associated life experiences appear to generally reinforce and
show “corresponsive” causal relationships with one another
(Denissen, Zarrett, & Eccles, 2007; Roberts & Wood, 2006).
On the basis of these numerous considerations, we believe
that our base expectation should be that the correlational relationships identified here generally indicate causal effects
of FIs on the behavioral traits they are associated with. At
the very least, these represent a myriad of sensible “empirically informed hypotheses” regarding the various FIs that
100 Wood, Hensler Gardner, & Harms
may be further explored as sources of Big Five-related behavioral traits and trait perceptions.
Broader Implications of a Functionalist Understanding
of Trait Covariation
The framework we have illustrated is a fairly different approach to the issue of trait covariation than is typically seen
in personality psychology, where covariation is sometimes
considered to be “explained” by a small number of factors
identified in structural investigations (e.g., the Big Five or
HEXACO factors). We will thus continue by first discussing
how structural factors such as the Big Five might be considered as linking with the current functionalist understanding
of trait covariation. Then, we will discuss some implications
of this manner of understanding the sources of behavioral
trait levels for additional topics within personality psychology, such as trait stability and trait measurement.
Locating Structural Factors in the Personality System
Within the empirical illustrations, our goal was to argue
that functionalist or process units such as expectancies, selfefficacies, values and goals can be enlisted to explain the covariation of traits such as assertiveness and sociability. We
did not directly address the issue of how structural factors
such as the Big Five might link to this understanding of trait
covariation. We consider four different perspectives which
may be compatible with the empirical illustrations.
Structural factors as approximating “distal causes” of
process variables. A first manner in which we might try to
locate structural factors in a functionalist framework is as
major causes of FIs and process variables. This view understands structural factors as approximating more distal causes
of FIs or process variables, which in turn mediate the effects
of structural factors on behavior.
This appears to be a fairly common understanding of
structural factors, as evidenced by how relations between
structural factors and behavior are often described in the
literature. To illustrate with recent examples: Duckitt and
Sibley (2009) considered the belief that some groups are superior to others as “ deriving directly from the personality dimension of Tough versus Tendermindedness (in Big-Five
terms, low Agreeableness)” (p. 102); Chan and Drasgow
(2001) argued that Big Five traits “relate to leader behaviors
through the individual’s motivation to lead, which in turn
affects the individual’s participation in leadership roles ”
(p. 481; also Judge, Piccolo, & Kosalka, 2009). Most directly,
Terracciano and McCrae (2012) argued that “it is perfectly
reasonable to say that party going is caused (proximally) by
liking people and that it is caused (distally) by extraversion”
(p. 449; italics added in examples above). This view is particularly reflected within Five Factor Theory, where process
variables are considered “characteristic adaptations” shaped
largely by the five factors measured by the NEO-PI, which
are considered “basic tendencies” (see McCrae & Costa, 2008,
Figure 5.1).
An appealing aspect of this conception of structural factors is that it provides some suggestions regarding how levels of FIs or process variables are themselves shaped. However, there are considerable questions as to the specific
nature of these more distal causes. One possibility is that
structural factors might approximate set-points or sources of
in
Psychological Review 122 (2015)
equilibrium within the personality system. As suggested by
DeYoung (2014) structural factors such as the Big Five may
approximate “persistent attractor states . . . [which] indicate
states toward which the person will tend to gravitate” (p. 3;
McCrae, 2009). Like an individual’s center of gravity, this
may not be a property of the individual we can hope to directly observe, but which helps to understand the individual’s behavior. As more distal sources of behavior, we may
however suspect structural factors as having stronger associations with particular biological systems than FIs. For instance, DeYoung (2006) has suggested that the plasticity and
stability “metatraits” that may exist above the Big Five may
have substantial correspondence to the dopaminergic or serotonergic systems, respectively.
Structural factors as “causally efficacious composites.”
A second perspective is suggested by Meehl’s (1993) understanding that “a mathematical factor can correspond to a
causally efficacious composite whose elements are qualitatively unlike” (p. 4). In other words, a structural factor may
not correspond to a particularly unitary thing, but rather be
better interpreted as something like “the sum of the diverse
things which cause the measured traits to covary.”
For instance, a “plasticity” or “Beta” dimension is sometimes thought to underlie the covariation between the extraversion and openness/intellect factors (DeYoung, 2006; Digman, 1997; Markon, Krueger, & Watson, 2005; but see Ashton
et al., 2009). However, if we examine the specific content of
IPIP items that have the highest loadings on this broader
Plasticity factor, as reported in DeYoung, 2010, Table 1, we
see that many of these are FIs of the sort considered here. (Indeed, because both our list of FIs in Study 2 and DeYoung’s,
2010 list of plasticity and stability indicators are IPIP items,
some of these are exactly the same items.) For instance, some
items concern descriptions of social skills (e.g., “have a natural talent for influencing people;” “express myself easily”),
others concern the desire for attention (e.g., “don’t mind being the center of attention”) and others an interest in self-development (e.g., “look forward to the opportunity to learn
and grow”). From this perspective, the diverse FIs found
to load most on this factor may not be indicators of some
deeper plasticity, but may instead indicate some of the more
important, relatively distinct processes that cause the associated traits to covary. In fact, it may be in part the fact that
qualitatively different processes such as these explain independent parts of the covariation of the associated traits which
causes these processes to receive high factor loadings, analogous to how variables retain nonzero weights in a simultaneous regression. More generally, any FI or process variable that is strongly associated with a structural factor should
usually be expected to be important for understanding why
the narrower behavioral traits and trait perceptions associated with the factor covary.
From this perspective, we could consider structural analyses as guiding us to the locations of some of the specific FIs
or process variables that are most important to explaining the
observed covariation between traits. A functionalist approach
such as that illustrated here might then be considered as decomposing these “causally efficacious composites” into their
more specific constituent elements (Johnson, 1999; Roberts,
2009). A couple advantages should follow from this, which
we can see in our empirical illustrations. First, we may see
more clearly that FIs such as these and others contribute to the
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation covariation of traits within a given domain relatively independently of one another. Additionally, by measuring these FIs
separately rather than aggregating them into such a composite, we can see more clearly that they have distinct effects on
traits in other domains. To use two of the characteristics just
listed, we see that people who “look forward to the opportunity to learn and grow” (#69 in Table 4) and “like to attract
attention” (#12) tend to be higher on both extraversion and
openness-related traits, as we should expect given that these
IPIP items have among the highest loadings on a broader plasticity factor (DeYoung, 2010; Table 1). However, these two FIs
have slightly different effects outside of these trait domains—
specifically, “looking forward to learning/growth opportunities” showed positive associations with emotional stability,
kindness, and practicality, whereas “liking attention” showed
negative associations with the same traits. These secondary
relationships are both sensible and valuable to know, but are
also largely masked when aggregated with many other items
to measure a broader plasticity factor.
Structural factors as useful summaries. Another interpretation of structural factors is detailed by Ashton and Lee
(2005), who argue that:
We can explain decisiveness, cautiousness, studiousness, and
practicality as manifestations of a more general construct of
‘Conscientiousness’ . . . [but this] should not be misconstrued
as a claim that these factors are causes of those characteristics. That is, although we explain these characteristics as examples of the broader constructs represented by the factors,
we do not explain the characteristics as consequences of those
broader constructs” (p. 15; authors’ italics).
Structural factors are understood more explicitly as descriptive from this perspective than those described previously. Consequently, the explanatory power of invoking structural factors to explain specific traits is understood as roughly
analogous to Funder’s (1991) understanding of the explanatory power of using trait attributions of specific behaviors,
who notes “such an explanation relates a specific behavioral
observation to a complex and general pattern of behavior” and
in this manner serve as useful “stopping places in the explanatory regress” (p. 36). For instance, if we say “Ramona is organized because she is high on conscientiousness,” we are in
effect saying “Ramona is organized because she is also cautious, studious, and practical,” and we are thus learning that
whatever processes commonly facilitate traits in the conscientiousness domain may be driving her organized behavior
(Ashton & Lee, 2005; McCrae & Costa, 1995). Traits covary in
part because they share environmental, biological, and functional pathways, and as structural factors are effective summaries of covariation, it may be prudent to consider the traits
at the level of structural factors rather than the narrower traits
in order to increase parsimony (Krueger, DeYoung, & Markon, 2010; Krueger & Markon, 2006). This resembles the most
recent position of McCrae (2014), who suggests that viewing a
structural factor as the sum (rather than the cause) of the narrower facets, much as shown in Figures 1B and 2, “is most certainly not meaningless; it should allow inferences about other,
extratest manifestations and correlates,” and elaborates that if
the various facets relate to other outcomes in similar ways, the
“facets may be legitimately combined because they are functionally equivalent” (p. 12).
Structural factors as particularly socially consequential traits. Finally, structural factors might be regarded as
pieces of information which are particularly useful or important
101
for perceivers to know about someone (Buss, 1996; Goldberg,
1981; McAdams & Pals, 2006; Srivastava, 2010). Interestingly,
this perspective understands structural factors as summaries rather than as approximating sources of the individual’s
behavior, but emphasizes that they may also be among the
most important sources of others’ behavior toward the individual. This view is depicted graphically in Figure 2.
This understanding of the Big Five and HEXACO factors rests on the fact that lexical studies involve the analysis of the most frequently used person descriptors in a given
culture. From this perspective, the reason communal structural factors (e.g., agreeableness, conscientiousness) emerge
prominently in lexical studies around the world (Ashton &
Lee, 2007; Saucier & Goldberg, 2001) is because these correlated tendencies are extremely consequential to a range of
relationships in any human society, and members of a culture thus develop and use many words to describe them. Estimates of communal factors more generally indicate the extent to which individuals will prioritize their interests over
the interests of others (Ashton & Lee, 2007; Wood, Tov, &
Costello, in press), and this is extremely useful information
to know about someone, regardless of the myriad processes
which may cause the person to do so (Buss, 2008; Cottrell,
Neuberg, & Li, 2007; Fiske, Cuddy, & Glick, 2007). In contrast, we have few words to describe tendencies to do activities left-handed versus right-handed (e.g., open doors, write,
hold mugs, throw balls), and thus don’t have a “handedness” factor in our trait structures. This is not because there
is any more or less covariation between these behavioral
tendencies, but because they are much less socially consequential, and thus less represented in lexical trait structures
(Goldberg & Saucier, 1995).
As support for this interpretation, Wood (in press) found
that trait terms with higher loadings on the Big Five dimensions had a larger impact on others’ relationship decisions
(e.g., whether to date or befriend someone) than terms with
negligible loadings. Similarly, conscientiousness-related
traits influence whether employers will hire, fire, or promote you, which is decidedly appropriate, as these behavioral traits impact the overall productivity of their company
(Barrick & Mount, 1991; Dudley, Orvis, Lebiecki, & Cortina,
2006). Moreover, conscientiousness-related traits impact relationship satisfaction and longevity (Bogg & Roberts, 2004;
Ozer & Benet-Martínez, 2006; Roberts, Kuncel, Shiner, Caspi,
& Goldberg, 2007). These are also fairly clearly causal effects: if an individual’s rate of conscientious behavior were
to become higher, his or her spouse’s satisfaction, organization’s earnings, job standing, and own life expectancy would
be expected to increase. Levels of conscientiousness-related
traits could increase by impacting a range of distinct FIs—
one individual by becoming more concerned with making
coworkers happy; a second by enjoying work more; a third
by becoming more able to filter out distractions—and we
might expect largely the same salutary effects. Their employers may in essence say “I don’t care why this employee behaves more conscientiously now, but this employee should
receive a promotion.” As depicted in Figure 2, it will often
be an individual’s behavioral tendencies which are the proximal causes of others’ responses to the individual, not the
processes that shaped them. The value of structural factors
such as the Big Five or HEXACO factors as a level of analysis may principally be their ability to summarize similarities
102 Wood, Hensler Gardner, & Harms
in how an individual’s traits tend to impact his or her interpersonal environment.
Jointly considering these perspectives.The four perspectives discussed above represent an incomplete survey
of views regarding how structural factor and functional explanations of behavior might be connected. However, there
is little reason to think they need to be mutually exclusive.
For instance, the axes of factors such as the Big Five will
gravitate toward items that have the most correlations with
other items. However, there are many reasons that a particular item will covary with many items in the set. It may do so
because it closely indicates a variable which influences many
conceptually different variables in the set. For instance, being secure in one’s self-worth (as indicated by adjectives like
well-adjusted and secure, and related to #4, #41, and #42 in Table 4) likely commonly facilitates extraverted, agreeable, conscientious, emotionally stable, and open behaviors. Alternatively, an item may covary with many other items by being
influenced by a qualitatively distinct variable that influences
many others in the set but is not itself closely indicated in
the analysis (e.g., functioning of the dopamine system, current mood, a response set). Alternatively, a particular item
will correlate with more items in a lexical set if many other
conceptually similar items have been frontloaded into the set
(e.g., many terms about warmth were entered into the analysis and few about handedness; Saucier & Goldberg, 1996).
We have already described our preferred understanding
of structural factors, which understands structural factors
first and foremost as indicating the most socially valuable
and consequential information about an individual’s behavioral tendencies, as depicted in Figure 2. However, even
without shifting from this conception of structural factors,
we can note that the most valuable traits for perceivers to
learn about a person should not simply be the person’s most
consequential behavioral tendencies, but the FIs or processes
that lead the person to perform them. For instance, knowing
whether an individual typically completes obligations and
works hard (i.e., acts conscientiously) may be of great value
to perceivers, but knowing whether the individual possesses
self-regulatory skills and feels it is important to honor commitments should be particularly useful toward predicting
such behaviors. For these and other reasons, structural analyses should simultaneously help to identify both the behavioral tendencies that are of particular interest to perceivers
and the FIs that are particularly important in shaping them.
More generally, structural factors such as the Big Five
may simultaneously indicate both the most important sources
and outputs of the individual’s personality system (Fleeson,
2012; McCabe & Fleeson, 2012). This could occur if the words
individuals use most frequently to describe one another follow a folk wisdom of encapsulating particularly important
stages of an individual’s goal-directed action patterns, such
as how they plan or initiate behaviors (DeYoung, 2014; Van
Egeren, 2009). This is an interesting possibility to be evaluated more fully in future research.
The Value of Greater Disaggregation in Personality
Measurement
Despite a general understanding that suppression effects
are rare and difficult to find (e.g., Paulhus, Robins, Trzesniewski, & Tracy, 2004), our results suggest that suppression effects are likely quite common: behavioral traits and
in
Psychological Review 122 (2015)
abstract trait perceptions are regularly uncorrelated in part
due to having some processes which increase their correlation and others that decrease their correlation in roughly
equal magnitude. Additionally, the model described here
and illustrated in Figure 2 indicates that FIs or behavioral
traits should not be considered manifestations of more basic factors such as extraversion. If structural factors do not
approximate the most important common causes of trait covariation, the rationale for aggregating narrower traits to
estimate them becomes somewhat weaker. The common
practice of aggregating FIs and process variables with the
behaviors they are most associated with makes it much more
difficult to see the diverse effects a single FI regularly has on
other behavioral traits. For instance, the common practice of
aggregating items indicating orderliness concerns, orderliness skills, and orderly behavior to estimate an individual’s
broader orderliness or conscientiousness, would have made
it more difficult to document the opposing effects of orderliness concerns and orderliness skills on anxiety and irritability levels that we found in both studies.
More generally, although factors such as the Big Five
represent useful ways of summarizing an individual’s traits
which impact other outcomes in similar ways, it should be a
rare circumstance when all the traits used to identify a structural factor are equally related with any particular outcome
of interest. Consequently, the practice of aggregating a large
number of correlated traits into an estimate of a single structural factor prior to explore their correlations with other variables of interest will generally result in a considerable loss of
information that is reliable and theoretically meaningful (McCrae, 2014). For instance, although extraversion measures are
sometimes regarded as having negligible or inconsistent relationships with gender (Feingold, 1994; Lynn & Martin, 1997),
narrower traits in the domain of extraversion have stronger
and more consistent relationships: measures related to assertiveness tends to be higher among men, whereas measures
related to warmth and positive affectivity tend to be higher
among women (Costa, Terracciano, & McCrae, 2001; Wood,
Nye, & Saucier, 2010).
Given the frequency with which structural factors have
been the preferred level of analysis in personality psychology, it is still surprisingly easy to make important contributions to the field by simply showing how very basic variables
(e.g., gender, age, political orientation, health, popularity,
intelligence, attractiveness) are associated with personality
traits measured at a more fine-grained level of analysis. For
instance, Soto, John, Gosling, and Potter (2011) recently demonstrated that simply shifting the level of analysis from the
broader Big Five factors to 10 narrower dimensions considerably refined the picture of how personality changes with
age as reported elsewhere (Roberts, Walton, & Viechtbauer,
2006; Srivastava, John, Gosling, & Potter, 2003). Although levels of the broad structural factor conscientiousness increase
considerably in adulthood, these increases seem to be largely
localized to traits related to self-discipline, whereas traits related to orderliness seem to increase negligibly (see also Jackson et al., 2009).
How Are Levels of Functionality Indicators Shaped?
As we have argued, FIs should be expected to be true
causes of behavioral trait levels, in that individuals should
gravitate toward behavioral trait levels that they experience as
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation having greater functionality. Consequently, much of the value
of identifying important FIs is in giving us a list of variables
we should target to influence behavioral trait levels. However,
we have mostly not discussed the issue of how these FIs might
be influenced. Additionally, we have focused principally on
explaining covariation of behavioral traits, but it is clear that
conceptually distinct FIs or process variables show considerable correlations with one another (e.g., positive social expectations, positive social skills, and liking social interactions).
In the same way that a single behavioral trait should be
understood as being shaped by several distinct FIs, a single
FI should be understood as being shaped by a large number
of distinct forces. Some of these will include predictable influences of FIs on one another; for instance, if an individual
increases in their desire to attain a particular goal, the individual can be expected to increase in their ability to do so,
and vice versa (Bandura, 1986; Denissen et al., 2007; Lent,
Brown, & Hackett, 1994). The causal effects of the ability,
expectancy, and valuation trait classes on one another can
be understood through the same overarching functionality
assumption that links these three trait classes to behavioral
traits (Wood & Denissen, in press). There should be pressure
for any trait to increase—whether it is a behavioral trait or an
FI—when the actor experiences a higher level of this trait to
be functional. For instance, it is generally functional to decrease one’s valuation of tasks you are unable to perform
(Baltes, 2003; Heckhausen, Wrosch, & Schulz, 2010), and to
find it more difficult or taxing to perform actions that you
find unrewarding (Kurzban, Duckworth, Kable, & Myers,
2013). Such functional dynamics will frequently lead conceptually distinct FIs to be correlated with one another and consequently can help to explain some of the syndromal qualities found in personality networks.
Additionally, an individual’s levels of FIs should be
shaped by features of the individual’s biology and culture
(Mischel & Shoda, 1995), or of the objective situation more
generally (Reis, 2008). For instance, an individual’s general
tendency to see other people positively, which is likely involved in influencing levels of many distinct behavioral traits,
might itself be influenced in predictable ways by environmental and biological factors ranging from physical attractiveness (Langlois et al., 2000), to the level of amygdala activation to happy faces (Omura, Todd Constable, & Canli, 2005),
the number of recent positive or negative events (Vaidya,
Gray, Haig, & Watson, 2002), and the degree of parental affection (Collins & Read, 1994), among many other things. As
with the effects of FIs on one another, the effects of biological and experiential variables on FIs can be understood as being mediated by altering the functionality of the FI itself. For
instance, it is somewhat more functional to perceive others
negatively if local rates of infectious disease are high (Schaller
& Murray, 2010), and children are more likely to develop intellectual abilities when in familial or cultural environments
that place greater value on such skills (Nisbett et al., 2012).
Implications for Understanding Trait Change and
Stability
Sometimes the considerable heritability estimates and
rank-order stability estimates of structural factors (e.g.,
103
Bouchard, 2004; Roberts & DelVecchio, 2000), and indications
of relatively small environmental effects on the development
of broad personality factors (Neyer & Asendorpf, 2001; Neyer
& Lehnart, 2007; Specht, Egloff, & Schmukle, 2011) are offered
as evidence that such factors are basic tendencies that are relatively impervious to change (Eysenck, 1970; McCrae & Costa,
2008). However, there is actually surprisingly little evidence
that structural factors are either more heritable or more stable
than the narrower traits they summarize. Once a trait measure’s instability over very short periods of time (e.g., 1 or 2
weeks) is accounted for, narrow traits appear to show heritability and stability estimates that may be comparable with
broader structural factors (McCrae, Kurtz, Yamagata, & Terracciano, 2011, Table 2; Wood & Wortman, 2012). Nor is it
clear that higher levels of these characteristics necessarily indicate that they are more “basic.” For instance, Johnson, McGue, Krueger, and Bouchard (2004) reported the heritability
of marital status to be on par with the heritabilities of personality traits, despite the fact that one’s likelihood of marrying is undoubtedly influenced by a variety of distinct traits.
A functionalist understanding of trait variation and covariation suggests some reasons to be both more and less
optimistic about the possibility of large-scale personality
change. First, the apparent resistance of personality traits to
environmental influence has likely been artificially inflated
by the aggregation of behavioral traits with distinct etiologies into estimates of structural factors. For instance, a particular assertiveness training program could indeed result in increased levels of assertiveness, but have negligible effects on
other traits in the extraversion domain, such as positive affect
or activity level, and even negative effects on others such as
sociability. If evaluators of this program were to search for
effects of the intervention on “personality traits” at the level
of structural factors, this effect will be diluted if not washed
out entirely, and the researchers may be misled to the conclusion that the intervention “does not impact personality
traits” when there are in fact several reliable effects of the intervention on specific behavioral traits and FIs. As a single
Big Five scale is a summary of a broad range of distinct traits,
the entire scale score should not shift much from some experience unless the experience impacts most of the narrower
traits contained within the scale and in the same direction.
Measuring distinct behavioral traits separately should thus
be expected to result in more frequent observation of environmental effects on trait development.
However, the current findings also suggest at least two
reasons to be more pessimistic about the possibility of largescale changes in behavioral traits. First, although the current
findings suggest some of the FIs or process variables we may
target to increase levels of the behavioral trait, levels of these
FIs will not be infinitely malleable. Rather, each is certain to
be highly influenced by numerous biological or environmental variables, which will often themselves be very stable over
time. As one example, many of the FIs shown in Tables 2 and
4 show considerable associations with self-reported sex. Females reported greater skill at understanding others’ feelings (#23 in Table 4, Cohen’s d = .72) than males, and lower
tendency to “love a good fight” (#26, d = −.49). As an individual’s biological sex impacts a number of FIs and doesn’t
2. This adjustment is necessary to adjust for the fact that heritability and longitudinal stability estimates will be higher for scales with more items,
all other things being equal. Further, as reviewed by McCrae et al. (2011), short test–retest stability (or “dependability”) estimates are likely more
conceptually appropriate corrections for unreliability than interitem reliabilities.
104 Wood, Hensler Gardner, & Harms
change, it will effectively function as a consistent force on the
associated FIs over time (mediated more proximally by its
effects on testosterone, gender role expectations, and so on).
Although this is an extreme case, many other biological and
environmental variables (e.g., weight, brain integrity, physical attractiveness, regional culture, social class) will be sufficiently stable over time to serve similar anchoring functions
for the FIs that influence behavioral traits.
Second, although these findings illuminate many of the
functional levers that might be pushed or pulled to influence behavioral trait levels, a second reason to be pessimistic about the possibility of large-scale changes is that there
are so many of them. In Tables 2 and 4 and the surrounding
discussion, we describe many of the distinct FIs that likely
contribute relatively independently to behavioral trait levels. And as we have noted, there will be numerous distinct
biological and environmental features involved in calibrating levels of any particular FI. When there are many distinct
determinants of a single behavior, the magnitude of the effect expected to occur by changing any one of them should
be small (Ahadi & Diener, 1989). For instance, we might expect that an individual who has suddenly inculcated a desire to become more sociable might observe little change in
his or her levels of sociability if this change is not also accompanied by some combination of greater expectations of
having positive social interactions, newfound social skills,
enjoyment of attention and social interactions, or interest in
others’ perspectives.
The Issue of Parsimony
The functional conception of the sources of trait covariation and the exploratory goals of our present empirical illustrations together resulted in our unusual decision to aggregate items as minimally as possible (see also Cramer et al.,
2012, 2010). We believe this to be a logical extension of functionalist frameworks, in that these understand trait levels
to be shaped by the value of their expected effects (Feather,
1982; Thorndike, 1913), and understand any given trait to
have a wide range of effects on the environment (e.g., Kruglanski et al., 2002). Additionally, we believe it has been valuable in providing within Tables 2 and 4 a much elaborated
understanding of how levels of Big Five-related traits tend
to relate to standard process variables such as self-efficacies,
expectancies, goals, and contingencies of liking and anxiety.
This in turn should help deepen the continuing integration of
prominent trait and process approaches to personality (Fleeson, 2012; Funder, 2001; Mischel, 2009).
However, it may be argued that this approach to covariation is not as parsimonious as explanations involving structural factors. To illustrate: Whereas our findings suggest that
the covariation of assertiveness and sociability is explained
by at least half a dozen relatively distinct FIs or more, the
Five Factor Model purports to explain the covariation of assertiveness, sociability, and at least four other distinct traits
(activity, excitement-seeking, positive emotionality, and
warmth) largely via the single variable of extraversion (McCrae & Costa, 2008). This type of reasoning is described by
Krueger, DeYoung, and Markon (2010) in their critique of a
network model recently proposed by Cramer et al. (2012),
which is similar in terms of its item-level analysis strategy
and very large number of distinct components:
in
Psychological Review 122 (2015)
With regard to model fit, the network models proposed by Cramer et al. are unlikely to emerge as optimal models when compared to latent variable [that is, structural factor] models. This is
because the network models fit by Cramer et al. contain a multitude of parameters; they are lacking in parsimony when compared with latent variable models. Models lacking in parsimony
often provide a poor relative fit to data and are thereby lower
in scientific utility because they amount to little more than reexpressions of observed data (p. 164).
We might agree with Krueger et al. that structural factor explanations of trait covariation look much more parsimonious
than network or functional explanations. For instance, compare Figure 2 of Krueger and Markon (2006) with our Figure
2, or with Cramer et al.’s Figure 2. As Krueger and colleagues
note, standard psychometric tools for judging relative parsimony (e.g., goodness-of-fit statistics) will almost undoubtedly
judge most structural models to be more parsimonious than
models such as what we have described here. But are they?
As noted by Dawkins (2005) “parsimony is always in the forefront of a scientist’s mind when choosing between theories,
but it isn’t always obvious how to judge it” (p. 118).
How parsimonious are structural explanations? As we
have reviewed, the level of parsimony associated with structural factor explanations depends in large part in how we
should regard the nature of the relationship between structural factors and behavior. If structural factors such as extraversion truly approximate the most important distal sources
of FIs or process variables, then explanations of covariation
by structural factors might be considered very parsimonious indeed. However, as we have discussed, there are many
perspectives regarding what structural factors actually correspond to.
Interestingly, theorists sometimes suggest that structural
factors correspond roughly to important FIs or process variables. For instance, an individual’s level of extraversion may
correspond highly to his or her valuation of reward stimuli
(Ashton, Lee, & Paunonen, 2002; Denissen & Penke, 2008;
Watson & Clark, 1997), which in turn might be associated
with the dopaminergic system (e.g., DeYoung, 2010). Alternatively, such factors might be best viewed as “causally efficacious composites” in which several fairly distinct FIs contribute independently to the explanation of covariation. In
both of these scenarios, it is ultimately FIs which are explaining trait covariation (e.g., the extravert is going to parties
because she more highly values social experiences). In this
sense, structural analyses may point the way to FIs that is
particularly important to explaining trait covariation, rather
than point to a distinct sort of explanatory variable.
The level of parsimony afforded by structural factor explanations of covariation is of course even more dubious if
we consider an individual’s level of a structural factor to
be primarily a summary of the indicated traits. As noted
by Borsboom and colleagues (Borsboom, Kievit, Cervone, &
Hood, 2009; Cramer et al., 2012), if considered in this fashion
an individual’s level of some structural factor like Extraversion should be regarded as caused by levels of more specific
tendencies to be assertive, sociable, and so on—in the sense
of being a mathematical summary—and it is simply a category error to “explain” covariation by the structural factor.
Given the ambiguity of what structural factors are, and
what they contribute to understanding behavior beyond
functional or process variables, we would argue that it may
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation be functionalist approaches which eschew any distinct role to
structural factors as sources of an individual’s more specific
traits that are actually more parsimonious. Granted, this is a
conceptual parsimony rather than a methodological one; there
will undoubtedly be considerable growing pains as personality psychologists and methodologists work to develop better methods for representing the complex and massively interacting networks of FIs or process variables that underlie
an individual’s behaviors. However, this is work worth doing, as understanding behavioral traits as being shaped by a
large number of FIs rather than by a small number of structural factors could reflect a more accurate of the causal dynamics that underlie variation and covariation (Borsboom et
al., 2009; Buss, 2008; Fleeson, 2012; Kurzban & Aktipis, 2007;
Mischel & Shoda, 1995). Consequently, the upshot of learning to deal with this complexity will be a better understanding of what traits are, and how trait levels are shaped. As
noted by Minsky (2007):
It might appear to make everything worse, to change some
things that looked simple at first into problems that now seem
more difficult. However, on a larger scale, this increase in
complexity will actually make our job easier. For, once we
split each old mystery into parts, we will have replaced each
old, big problem with several new and smaller ones—each of
which may still be hard but no longer seem unsolvable. (p. 2)
We believe that the large role structural factors have
played in explanations of trait covariation may have in some
ways served to slow our understanding of this phenomenon. We may need to become more comfortable using a wide
number of narrower process variables to explain trait covariation. Decreasing the role of structural factors could help
to remove a large cognitive and theoretical impediment to
a more accurate understanding of trait covariation, similar
to eliminating consideration of a singular homunculus as a
source of our experience of consciousness in favor of models
favoring a much larger network of narrower processes (Dennett, 1993; Kurzban & Aktipis, 2007; Minsky, 1988).
Placing some limits on explanatory complexity. Functionalist frameworks to covariation are consistent with network approaches to understanding personality functioning that accommodate hundreds of interacting components
(Cramer et al., 2012; Mischel & Shoda, 1995). Each distinct
effect that might result from behavior, and separate valuation of each effect, can be represented as a separate node
in a personality network, and many of these should be expected to explain independent variance in a given behavioral trait. As we have shown (e.g., Figure 2), this can result
in a considerably more complex account of the sources of
trait covariation than those found in a standard structural
approaches.
At the same time, recommending that a small number of
structural factors be replaced with hundreds or even thousands of items or variables to understand trait covariation
can be meaningfully compared with recommending we aspire for the “great blooming, buzzing confusion” that James
(1890, p. 488) described as an infant’s phenomenological experience before developing useful concepts for representing
the world. This is not ultimately what we are recommending.
Rather, as in the development of expertise (Ericsson & Lehmann, 1996), achieving a mature and useful understanding
of a phenomenon involves reducing many molecular pieces
105
of information into more molar “chunks,” and drawing out
more general principles (Funder, 1991).
It is thus useful to briefly review some of the various steps
that were taken here to reduce explanatory complexity at the
level of FIs (see also S1 in the Supplementary Materials). We
began with initial sets of 463 and 1,351 items in Studies 1 and
2, respectively, which were correlated with measures of Big
Five-related trait perceptions. However, cluster analysis was
then used to group items that (a) related similarly to one another, and (b) related similarly to trait perceptions, and reduce these to a single row in Tables 2 and 4. For instance,
“concern for order” and “organization skills” are highly correlated with one another, but these distinctions were teased
apart in both studies by relating differently to anxiety-related
traits. However, distinctions such as “feeling uncomfortable
around disorder” and “feeling stressed when there is not a
plan” were initially measured as separate items, but were
collapsed by the same considerations. Interestingly, these
steps paralleled some of those that were used in the development of Block’s diverse Q-sort instruments (see Block, 1961).
As in that context, these steps lend themselves to a greater
number of scales or components than reduction techniques
like factor analysis which mainly prioritize interitem correlations, but also provide an effective limit to the level of explanatory complexity that will be tolerated in the analysis.
At a more theoretical level, a functionalist framework further reduces explanatory complexity in a couple ways. First,
it provides a classification scheme for different aspects of a
personality system. Rather than considering each item separately (Cramer et al., 2012), these can be conceptually classified into the three classes of FIs—abilities/efficacies, expectancies, and valuations—or as behavioral traits and more
abstract trait perceptions. These trait classes can be used to
reliably classify items in personality scales (e.g., “I like . . .”
statements are valuation indicators; “I easily . . .” statements
are ability indicators). More importantly, the way in which
FIs should relate to behavioral patterns are articulated in
a range of expectancy-value, rational actor, and economic
models (e.g., Almlund et al., 2011; Feather, 1982; Gintis, 2007;
Vroom, 1964). Specifically, FIs should influence behavioral
trait levels principally by influencing their functionality. This
overarching idea, which we have referred to as the functionally assumption, serves as a parsimonious theoretical guide as
to when we should expect to observe associations between
FIs and behavioral traits (or between nodes in a personality
network) and when we should not.
We believe this assumption holds up extremely well in
our empirical illustrations. Specifically, it is difficult to identify clear cases within the hundreds of correlations listed in
Tables 2 and 4 where behavioral traits were higher among
individuals who found it more effortful or difficult to perform trait-identifying behaviors, or among those who had
a greater expectation of negative outcomes of such behaviors. Rather, we suspect the specific positive, negative, and
nonsignificant associations observed between FIs and behavioral traits generally appear sensible and nonsurprising to
most readers, due mainly to their own fairly intuitive application of this functionalist assumption. For instance, to
predict sociable behavioral tendencies to be positively correlated with the expectation that others are unkind would
likely strike most people as absurd, principally because it
106 Wood, Hensler Gardner, & Harms
typically becomes less (rather than more) functional to socialize if one has such expectations. More generally, whereas
both network and structural approaches to covariation are
at their heart mathematical modeling techniques, functionalist approaches at their heart suggest how the content of
such networks should be organized. As more work is done
to detail how this functionality assumption can be more explicitly formalized, we expect that it will serve as a powerful guide for understanding the structure of complex personality networks.
Conclusion
The fact that distinct personality traits covary with one
another has been suggested as a major rationale for casting
structural factors such as the Big Five or HEXACO factors as
“underlying” or “explaining” trait covariation (Cattell, 1950;
McCrae & Costa, 1995; Tellegen, 1991). However, such explanations are sometimes considered problematic by proponents of process approaches, due in part to concerns of circularity and the question of how structural factors should be
operationalized outside of their indicators (Bandura, 1999;
Cervone, 1999; Cramer et al., 2012; Mischel, 1968). We illustrate here how the ability/efficacy, expectancy, and valuation trait classes common to a range of functionalist and process approaches—which we have considered here primarily
as functionality indicators or FIs—can be enlisted to explain
trait covariation without necessarily positing a distinct role
for structural factors.
As we have detailed here, explanations of behavioral
trait covariation through process variables or FIs may offer a particularly valuable method for understanding trait
covariation. Relative to explanations of covariation that enlist structural factors, such explanations more directly provide a specific process understanding of where behavioral
trait levels come from, and of how (and perhaps how much)
they might be altered. Such a framework should thus help
us shed new light not just on the specific sources of trait covariation, but on many other issues concerning the nature
of personality traits.
References
Ahadi, S., & Diener, E. (1989). Multiple determinants and effect
size. Journal of Personality and Social Psychology, 56, 398– 406.
doi: 10.1037/0022-3514.56.3.398
Ajzen, I. (1991). The theory of planned behavior. Organizational
Behavior and Human Decision Processes, 50, 179– 211. doi:
10.1016/0749-5978(91)90020-T
Allport, G. W. (1937). Personality: A psychological interpretation. Oxford, UK: Holt.
Almlund, M., Duckworth, A. L., Heckman, J. J., & Kautz, T.
D. (2011). Personality psychology and economics. In E.
A.Hanushek, S.Machin, & L.Woessmann (Eds.), Handbook of
the economics of education (pp. 1– 181). Amsterdam, The Netherlands: Elsevier. doi: 10.1016/B978-0-444-53444-6.00001-8
Ashton, M. C., & Lee, K. (2005). A defence of the lexical approach
to the study of personality structure. European Journal of Personality, 19, 5– 24. doi: 10.1002/per.541
Ashton, M. C., & Lee, K. (2007). Empirical, theoretical, and
practical advantages of the HEXACO model of personality
in
Psychological Review 122 (2015)
structure. Personality and Social Psychology Review, 11, 150– 166.
doi: 10.1177/1088868306294907
Ashton, M. C., Lee, K., Goldberg, L. R., & de Vries, R. E.
(2009). Higher order factors of personality: Do they
exist?Personality and Social Psychology Review, 13, 79– 91. doi:
10.1177/1088868309338467
Ashton, M. C., Lee, K., & Paunonen, S. V. (2002). What is the central feature of extraversion? Social attention versus reward
sensitivity. Journal of Personality and Social Psychology, 83, 245–
252. doi: 10.1037/0022-3514.83.1.245
Atkinson, J. W. (1957). Motivational determinants of risk-taking behavior. Psychological Review, 64, 359– 372. doi: 10.1037/
h0043445
Austin, J. T., & Vancouver, J. B. (1996). Goal constructs in psychology: Structure, process, and content. Psychological Bulletin, 120,
338– 375. doi: 10.1037/0033-2909.120.3.338
Baltes, P. B. (2003). On the incomplete architecture of human ontogeny: Selection, optimization and compensation as foundation of developmental theory. In U. M.Staudinger &
U.Lindenberger (Eds.), Understanding human development: Dialogues with lifespan psychology (pp. 17– 43). Amsterdam, The
Netherlands: Kluwer. doi: 10.1007/978-1-4615-0357-6_2
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of
behavioral change. Psychological Review, 84, 191– 215. doi:
10.1037/0033-295X.84.2.191
Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (1999). Social cognitive theory of personality. In L.
A.Pervin & O. P.John (Eds.), Handbook of personality: Theory and
research (2nd ed., pp. 154– 196). New York, NY: Guilford Press.
Barrick, M. R., & Mount, M. K. (1991). The Big Five personality
dimensions and job performance: A meta-analysis. Personnel
Psychology, 44, 1– 26. doi: 10.1111/j.1744-6570.1991.tb00688.x
Bertalanffy, L. V. (1968). General system theory: Foundations, development, applications. New York, NY: Braziller.
Block, J. (1961). The Q-sort method in personality assessment and
psychiatric research. Springfield, IL: Charles C. Thomas. doi:
10.1037/13141-000
Bogg, T., & Roberts, B. W. (2004). Conscientiousness and healthrelated behaviors: A meta-analysis of the leading behavioral
contributors to mortality. Psychological Bulletin, 130, 887– 919.
doi: 10.1037/0033-2909.130.6.887
Borsboom, D., Kievit, R. A., Cervone, D., & Hood, S. B. (2009).
The two disciplines of scientific psychology, or: The disunity of psychology as a working hypothesis. In J.Valsiner,
P. C. M.Molenaar, M. C. D. P.Lyra, & N.Chaudhary (Eds.),
Dynamic process methodology in the social and developmental sciences (pp. 67– 97). New York, NY: Springer. doi:
10.1007/978-0-387-95922-1_4
Bouchard, T. J. J. (2004). Genetic influence on human psychological traits: A survey. Current Directions in Psychological Science,
13, 148– 151. doi: 10.1111/j.0963-7214.2004.00295.x
Brehm, J. W., & Self, E. A. (1989). The intensity of motivation. Annual Review of Psychology, 40, 109– 131. doi: 10.1146/annurev.
ps.40.020189.000545
Buss, D. M. (1996). Social adaptation and five major factors of
personality. In J. S.Wiggins (Ed.), The Five-Factor model of personality: Theoretical perspectives (pp. 180– 207). New York, NY:
Guilford Press.
Buss, D. M. (2008). Human nature and individual differences: Evolution of human personality. In O. P.John, R. W.Robins, & L.
A.Pervin (Eds.), Handbook of personality: Theory and research (3rd
ed., pp. 29– 60). New York, NY: Guilford Press.
Buss, D. M., & Craik, K. H. (1983). The act frequency approach
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation to personality. Psychological Review, 90, 105– 126. doi:
10.1037/0033-295X.90.2.105
Buss, D. M., Gomes, M., Higgins, D. S., & Lauterbach, K. (1987).
Tactics of manipulation. Journal of Personality and Social Psychology, 52, 1219– 1229. doi: 10.1037/0022-3514.52.6.1219
Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior. New York, NY: Cambridge. doi: 10.1017/
CBO9781139174794
Caspi, A., Houts, R. M., Belsky, D. W., Goldman-Mellor, S. J., Harrington, H., Israel, S., . . .Moffitt, T. E. (2014). The p factor:
One general psychopathology factor in the structure of psychiatric disorders?Clinical Psychological Science, 2, 119– 137. doi:
10.1177/2167702613497473
Cattell, R. B. (1946). Description and measurement of personality. Oxford, UK: World Book Company.
Cattell, R. B. (1950). Describing personality through traits. In
R.Cattell (Ed.), Personality: A systematic theoretical and factual
study (1st ed., pp. 21– 44). New York, NY: McGraw-Hill.
Cervone, D. (1999). Bottom-up explanation in personality psychology: The case of cross-situational coherence. In D.Cervone &
Y.Shoda (Eds.), The coherence of personality: Social-cognitive bases
of consistency, variability, and organization (pp. 303– 341). New
York, NY: Guilford Press.
Chan, K.-Y., & Drasgow, F. (2001). Toward a theory of individual differences and leadership: Understanding the motivation to lead. Journal of Applied Psychology, 86, 481– 498. doi:
10.1037/0021-9010.86.3.481
Cialdini, R. B., & Trost, M. R. (1998). Social influence: Social
norms, conformity and compliance. In D. T.Gilbert, S. T.Fiske,
& G.Lindzey (Eds.), The handbook of social psychology (4th ed.,
Vols. 1 and 2, pp. 151– 192). New York, NY: McGraw-Hill.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd
ed.). Mahwah, NJ: Erlbaum.
Collins, N. L., & Read, S. J. (1994). Cognitive representations of attachment: The structure and function of working models. In
K.Bartholomew & D.Perlman (Eds.), Attachment processes in
adulthood (pp. 53– 90). London, UK: Jessica Kingsley.
Connor-Smith, J. K., & Flachsbart, C. (2007). Relations between personality and coping: A meta-analysis. Journal of Personality and Social Psychology, 93, 1080– 1107. doi:
10.1037/0022-3514.93.6.1080
Costa, P. T., Jr., Terracciano, A., & McCrae, R. R. (2001). Gender
differences in personality traits across cultures: Robust and
surprising findings. Journal of Personality and Social Psychology,
81, 322– 331. doi: 10.1037/0022-3514.81.2.322
Cottrell, C. A., Neuberg, S. L., & Li, N. P. (2007). What do people desire in others? A sociofunctional perspective on
the importance of different valued characteristics. Journal of Personality and Social Psychology, 92, 208– 231. doi:
10.1037/0022-3514.92.2.208
Cramer, A. O. J., Van der Sluis, S., Noordhof, A., Wichers, M., Geschwind, N., Aggen, S. H., . . .Borsboom, D. (2012). Dimensions of normal personality as networks in search of equilibrium: You can’t like parties if you don’t like people. European
Journal of Personality, 26, 414– 431. doi: 10.1002/per.1866
Cramer, A. O. J., Waldorp, L. J., van der Maas, H. L., & Borsboom, D. (2010). Comorbidity: A network perspective. Behavioral and Brain Sciences, 33((2–3)): 137– 150. doi: 10.1017/
S0140525X09991567
Dawkins, R. (2005). The ancestor’s tale: A pilgrimage to the dawn of
evolution. New York, NY: Houghton Mifflin.
Denissen, J. J. A., & Penke, L. (2008). Motivational individual reaction norms underlying the Five-Factor model of personality:
107
First steps towards a theory-based conceptual framework.
Journal of Research in Personality, 42, 1285– 1302. doi: 10.1016/j.
jrp.2008.04.002
Denissen, J. J. A., Zarrett, N. R., & Eccles, J. S. (2007). I like to do it,
I’m able, and I know I am: Longitudinal couplings between domain-specific achievement, self-concept, and interest. Child Development, 78, 430– 447. doi: 10.1111/j.1467-8624.2007.01007.x
Dennett, D. C. (1993). Consciousness explained. London, UK:
Penguin.
DeYoung, C. G. (2006). Higher-order factors of the Big Five in a
multi-informant sample. Journal of Personality and Social Psychology, 91, 1138– 1151. doi: 10.1037/0022-3514.91.6.1138
DeYoung, C. G. (2010). Toward a theory of the Big Five. Psychological Inquiry, 21, 26– 33. doi: 10.1080/10478401003648674
DeYoung, C. G. (2014). Cybernetic Big Five Theory. Journal of Research in Personality. Advance online publication.
DeYoung, C. G., Quilty, L. C., & Peterson, J. B. (2007). Between facets and domains: 10 aspects of the Big Five. Journal of Personality and Social Psychology, 93, 880– 896. doi:
10.1037/0022-3514.93.5.880
Digman, J. M. (1997). Higher-order factors of the Big Five. Journal of Personality and Social Psychology, 73, 1246– 1256. doi:
10.1037/0022-3514.73.6.1246
Duckitt, J., & Sibley, C. G. (2009). A dual-process motivational
model of ideology, politics, and prejudice. Psychological Inquiry, 20, 98– 109. doi: 10.1080/10478400903028540
Dudley, N. M., Orvis, K. A., Lebiecki, J. E., & Cortina, J. M. (2006).
A meta-analytic investigation of conscientiousness in the prediction of job performance: Examining the intercorrelations
and the incremental validity of narrow traits. Journal of Applied
Psychology, 91, 40– 57. doi: 10.1037/0021-9010.91.1.40
Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach
to motivation and personality. Psychological Review, 95, 256–
273. doi: 10.1037/0033-295X.95.2.256
Eccles, J. S., & Wigfield, A. (2002). Motivational beliefs, values, and
goals. Annual Review of Psychology, 53, 109– 132. doi: 10.1146/
annurev.psych.53.100901.135153
Ericsson, K. A., & Lehmann, A. C. (1996). Expert and exceptional performance: Evidence of maximal adaptation to task
constraints. Annual Review of Psychology, 47, 273– 305. doi:
10.1146/annurev.psych.47.1.273
Eysenck, H. J. (1970). The structure of human personality. New York,
NY: Methuen.
Feather, N. T. (1959). Subjective probability and decision under
uncertainty. Psychological Review, 66, 150– 164. doi: 10.1037/
h0045692
Feather, N. T. (1982). Expectations and actions: Expectancy-value models in psychology. Mahwah, NJ: Erlbaum.
Feingold, A. (1994). Gender differences in personality: A
meta-analysis. Psychological Bulletin, 116, 429– 456. doi:
10.1037/0033-2909.116.3.429
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison Wesley.
Fiske, S. T., Cuddy, A. J., & Glick, P. (2007). Universal dimensions
of social cognition: Warmth and competence. Trends in Cognitive Sciences, 11, 77– 83. doi: 10.1016/j.tics.2006.11.005
Fleeson, W. (2001). Toward a structure- and process-integrated
view of personality: Traits as density distribution of states.
Journal of Personality and Social Psychology, 80, 1011– 1027. doi:
10.1037/0022-3514.80.6.1011
Fleeson, W. (2012). Perspectives on the person: Rapid growth and
opportunities for integration. In K.Deaux & M.Snyder (Eds.),
108 Wood, Hensler Gardner, & Harms
The Oxford handbook of personality and social psychology (pp. 33–
63). New York, NY: Oxford.
Fleeson, W., & Jolley, S. (2006). A proposed theory of the adult
development of intraindividual variability in trait-manifesting behavior. In D. K.Mroczek & T. D.Little (Eds.), Handbook
of personality development (pp. 41– 59). Mahwah, NJ: Erlbaum.
Fleeson, W., Malanos, A. B., & Achille, N. M. (2002). An intraindividual process approach to the relationship between extraversion and positive affect: Is acting extraverted as “good” as
being extraverted?Journal of Personality and Social Psychology,
83, 1409– 1422. doi: 10.1037/0022-3514.83.6.1409
Funder, D. C. (1991). Global traits: A Neo-Allportian approach to
personality. Psychological Science, 2, 31– 39. doi: 10.1111/j.14679280.1991.tb00093.x
Funder, D. C. (2001). Personality. Annual Review of Psychology, 52,
197– 221. doi: 10.1146/annurev.psych.52.1.197
Gibson, J. (1977). The theory of affordances. In R.Shaw &
J.Bransford (Eds.), Perceiving, acting, and knowing (pp. 67– 82).
Hillsdale, NJ: Erlbaum.
Gigerenzer, G., & Goldstein, D. G. (1996). Reasoning the fast and
frugal way: Models of bounded rationality. Psychological Review, 103, 650– 669. doi: 10.1037/0033-295X.103.4.650
Gintis, H. (2007). A framework for the unification of the behavioral sciences. Behavioral and Brain Sciences, 30, 1– 16. doi:
10.1017/S0140525X07000581
Gintis, H. (2009). The bounds of reason: Game theory and the unification of the behavioral sciences. Princeton, NJ: Princeton University Press.
Goldberg, L. R. (1981). Language and individual differences: The
search for universals in personality lexicons. Review of Personality and Social Psychology, 2, 141– 165.
Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48, 26– 34. doi:
10.1037/0003-066X.48.1.26
Goldberg, L. R. (1999). A broad-bandwidth, public domain, personality inventory measuring the lower-level facets of several
five-factor models. Personality Psychology in Europe, 7, 7– 28.
Goldberg, L. R. (2008). The Eugene-Springfield community sample:
Information available from the research participants. Eugene, OR:
Oregon Research Institute.
Goldberg, L. R., Johnson, J. A., Eber, H. W., Hogan, R., Ashton,
M. C., Cloninger, C. R., & Gough, H. G. (2006). The International Personality Item Pool and the future of public-domain
personality measures. Journal of Research in Personality, 40, 84–
96. doi: 10.1016/j.jrp.2005.08.007
Goldberg, L. R., & Saucier, G. (1995). So what do you propose we
use instead? A reply to Block. Psychological Bulletin, 117, 221–
225. doi: 10.1037/0033-2909.117.2.221
Guilford, J. P. (1959). Personality. New York, NY: McGraw-Hill.
Harms, P. D., Roberts, B. W., & Wood, D. (2007). Who shall lead?
An integrative personality approach to the study of the antecedents of status in informal social organizations. Journal of Research in Personality, 41, 689– 699. doi: 10.1016/j.jrp.2006.08.001
Heckhausen, H. (1977). Achievement motivation and its constructs: A cognitive model. Motivation and Emotion, 1, 283–
329. doi: 10.1007/BF00992538
Heckhausen, J., Wrosch, C., & Schulz, R. (2010). A motivational
theory of life-span development. Psychological Review, 117, 32–
60. doi: 10.1037/a0017668
Higgins, E. T. (2000). Does personality provide unique explanations for behaviour? Personality as cross-person variability in general principles. European Journal of Personality, 14,
391– 406. doi: 10.1002/1099-0984(200009/10)14:5<391::AIDPER394>3.0.CO;2-6
in
Psychological Review 122 (2015)
Jackson, J. J., Bogg, T., Walton, K. E., Wood, D., Harms, P. D., LodiSmith, J., . . .Roberts, B. W. (2009). Not all conscientiousness
scales change alike: A multimethod, multisample study of age
differences in the facets of conscientiousness. Journal of Personality and Social Psychology, 96, 446– 459. doi: 10.1037/a0014156
James, W. (1890). The principles of psychology (Vol. I). New York,
NY: Holt. doi: 10.1037/11059-000
John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift
to the integrative Big Five trait taxonomy: History, measurement, and conceptual issues. In O. P.John, R. W.Robins, & L.
A.Pervin (Eds.), Handbook of personality: Theory and research (3rd
ed., pp. 114– 158). New York, NY: Guilford Press.
Johnson, J. A. (1999). Persons in situations: Distinguishing new wine from old wine in new bottles. European Journal of Personality, 13, 443– 453. doi: 10.1002/
(SICI)1099-0984(199909/10)13:5<443::AID-PER358>3.0.CO;2-9
Johnson, W., McGue, M., Krueger, R. F., & Bouchard, T. J. J., Jr.
(2004). Marriage and personality: A genetic analysis. Journal of Personality and Social Psychology, 86, 285– 294. doi:
10.1037/0022-3514.86.2.285
Judge, T. A., Piccolo, R. F., & Kosalka, T. (2009). The bright and
dark sides of leader traits: A review and theoretical extension
of the leader trait paradigm. The Leadership Quarterly, 20, 855–
875. doi: 10.1016/j.leaqua.2009.09.004
Kahneman, D. (2011). Thinking, fast and slow. New York, NY: Farrar, Straus & Giroux.
Katz, M. L., & Shapiro, C. (1985). Network externalities, competition, and compatibility. The American Economic Review, 75,
424– 440.
Kenrick, D. T., Griskevicius, V., Sundie, J. M., Li, N. P., Li, Y. J., &
Neuberg, S. L. (2009). Deep rationality: The evolutionary economics of decision making. Social Cognition, 27, 764– 785. doi:
10.1521/soco.2009.27.5.764
Krueger, R. F., Deyoung, C. G., & Markon, K. E. (2010). Toward
scientifically useful quantitative models of psychopathology:
The importance of a comparative approach. Behavioral and
Brain Sciences, 33, 163– 164. doi: 10.1017/S0140525X10000646
Krueger, R. F., & Markon, K. E. (2006). Reinterpreting comorbidity: A model-based approach to understanding and classifying
psychopathology. Annual Review of Clinical Psychology, 2, 111–
133. doi: 10.1146/annurev.clinpsy.2.022305.095213
Kruglanski, A. W., Shah, J. Y., Fishbach, A., Friedman, R., Chun,
W. Y., & Sleeth-Keppler, D. (2002). Advances in experimental social psychology. In M. P.Zanna (Ed.), A theory of goal systems (Vol. 34, pp. 331– 378). San Diego, CA: Academic Press.
Kurzban, R., & Aktipis, C. A. (2007). Modularity and the social
mind: Are psychologists too self-ish?Personality and Social Psychology Review, 11, 131– 149. doi: 10.1177/1088868306294906
Kurzban, R., Duckworth, A., Kable, J., & Myers, J. (2013). A cost/
benefit model of subjective effort and task performance. Behavioral and Brain Sciences. Advance online publication.
Kyl-Heku, L. M., & Buss, D. M. (1996). Tactics as units of analysis
in personality psychology: An illustration using tactics of hierarchy negotiation. Personality and Individual Differences, 21,
497– 517. doi: 10.1016/0191-8869(96)00103-1
Langlois, J. H., Kalakanis, L., Rubenstein, A. J., Larson, A., Hallam,
M., & Smoot, M. (2000). Maxims or myths of beauty? A metaanalytic and theoretical review. Psychological Bulletin, 126, 390–
423. doi: 10.1037/0033-2909.126.3.390
Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying
social cognitive theory of career and academic interest, choice, and
performance, 45, 79– 122.
Lewis, G. J., & Bates, T. C. (2014). How genes influence personality: Evidence from multi-facet twin analyses of the HEXACO
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation dimensions. Journal of Research in Personality, 51, 9– 17. doi:
10.1016/j.jrp.2014.04.004
Loehlin, J. C. (2004). Latent variable models: An introduction to factor, path, and structural equation analysis (4th ed.). Mahwah,
NJ: Erlbaum.
Lukaszewski, A. W. (2013). Testing an adaptationist theory of trait
covariation: Relative bargaining power as a common calibrator of an interpersonal syndrome. European Journal of Personality, 27, 328– 345. doi: 10.1002/per.1908
Lynn, R., & Martin, T. (1997). Gender differences in extraversion,
neuroticism, and psychoticism in 37 nations. The Journal of Social Psychology, 137, 369– 373. doi: 10.1080/00224549709595447
MacKinnon, D. P., Krull, J. L., & Lockwood, C. M. (2000). Equivalence of the mediation, confounding and suppression effect.
Prevention Science, 1, 173– 181. doi: 10.1023/A:1026595011371
Magidson, J. F., Roberts, B. W., Collado-Rodriguez, A., & Lejuez,
C. W. (2012). Theory-driven intervention for changing personality: Expectancy value theory, behavioral activation, and
Conscientiousness. Developmental Psychology, 50, 1442– 1450.
Malle, B. F. (1999). How people explain behavior: A new theoretical framework. Personality and Social Psychology Review, 3, 23–
48. doi: 10.1207/s15327957pspr0301_2
Markon, K. E., Krueger, R. F., & Watson, D. (2005). Delineating the
structure of normal and abnormal personality: An integrative
hierarchical approach. Journal of Personality and Social Psychology, 88, 139– 157. doi: 10.1037/0022-3514.88.1.139
McAdams, D. P., & Pals, J. L. (2006). A new Big Five: Fundamental principles for an integrative science of personality. American Psychologist, 61, 204– 217. doi: 10.1037/0003-066X.61.3.204
McCabe, K. O., & Fleeson, W. (2012). What is extraversion for? Integrating trait and motivational perspectives and identifying
the purpose of extraversion. Psychological Science, 23, 1498–
1505. doi: 10.1177/0956797612444904
McCrae, R. R. (2009). The physics and chemistry of personality. Theory & Psychology, 19, 670– 687. doi:
10.1177/0959354309341928
McCrae, R. R. (2014). A more nuanced view of reliability specificity in the trait hierarchy. Personality and Social Psychology Review. Advance online publication.
McCrae, R. R., & Costa, P. T. (1995). Trait explanations in personality psychology. European Journal of Personality, 9, 231– 252.
doi: 10.1002/per.2410090402
McCrae, R. R., & Costa, P. T. J. (2008). The five-factor theory of
personality. In O. P.John, R. W.Robins, & L. A.Pervin (Eds.),
Handbook of personality: Theory and research (3rd ed., pp. 159–
181). New York, NY: Guilford Press.
McCrae, R. R., Kurtz, J. E., Yamagata, S., & Terracciano, A. (2011).
Internal consistency, retest reliability, and their implications
for personality scale validity. Personality and Social Psychology
Review, 15, 28– 50. doi: 10.1177/1088868310366253
Meehl, P. E. (1993). Four queries about factor reality. History and
Philosophy of Psychology Bulletin, 5, 4– 5.
Meston, C. M., & Buss, D. M. (2007). Why humans have sex.
Archives of Sexual Behavior, 36, 477– 507. doi: 10.1007/
s10508-007-9175-2
Metcalfe, J., & Mischel, W. (1999). A hot/cool-system analysis of
delay of gratification: Dynamics of willpower. Psychological Review, 106, 3– 19. doi: 10.1037/0033-295X.106.1.3
Minsky, M. (1988). Society of mind. New York, NY: Simon &
Schuster.
Minsky, M. (2007). The emotion machine: Commonsense thinking, artificial intelligence, and the future of the human mind. New York,
NY: Simon & Schuster.
109
Mischel, W. (1968). Personality and assessment. Hoboken, NJ: Wiley.
Mischel, W. (2009). From Personality and Assessment (1968). to
personality science, 2009. Journal of Research in Personality, 43,
282– 290. doi: 10.1016/j.jrp.2008.12.037
Mischel, W., & Shoda, Y. (1995). A cognitive-affective system theory of personality: Reconceptualizing situations, dispositions,
dynamics, and invariance in personality structure. Psychological Review, 102, 246– 268. doi: 10.1037/0033-295X.102.2.246
Morf, C. C., & Rhodewalt, F. (2001). Unraveling the paradoxes of narcissism: A dynamic self-regulatory processing model. Psychological Inquiry, 12, 177– 196. doi: 10.1207/
S15327965PLI1204_1
Neyer, F. J., & Asendorpf, J. B. (2001). Personality–relationship
transaction in young adulthood. Journal of Personality and Social
Psychology, 81, 1190– 1204. doi: 10.1037/0022-3514.81.6.1190
Neyer, F. J., & Lehnart, J. (2007). Relationships matter in personality development: Evidence from an 8-year longitudinal study
across young adulthood. Journal of Personality, 75, 535– 568.
doi: 10.1111/j.1467-6494.2007.00448.x
Nisbett, R. E., Aronson, J., Blair, C., Dickens, W., Flynn, J., Halpern, D. F., & Turkheimer, E. (2012). Intelligence: New findings
and theoretical developments. American Psychologist, 67, 130–
159. doi: 10.1037/a0026699
Omura, K., Todd Constable, R., & Canli, T. (2005). Amygdala
gray matter concentration is associated with extraversion
and neuroticism. NeuroReport, 16, 1905– 1908. doi: 10.1097/01.
wnr.0000186596.64458.76
Ozer, D. J., & Benet-Martínez, V. (2006). Personality and the prediction of consequential outcomes. Annual Review of Psychology, 57, 401– 421. doi: 10.1146/annurev.psych.57.102904.190127
Ozer, D. J., & Reise, S. P. (1994). Personality assessment. Annual Review of Psychology, 45, 357– 388. doi: 10.1146/annurev.
ps.45.020194.002041
Paulhus, D. L., Robins, R. W., Trzesniewski, K. H., & Tracy, J.
L. (2004). Two replicable suppressor situations in personality research. Multivariate Behavioral Research, 39, 303– 328. doi:
10.1207/s15327906mbr3902_7
Reis, H. T. (2008). Reinvigorating the concept of situation in social
psychology. Personality and Social Psychology Review, 12, 311–
329. doi: 10.1177/1088868308321721
Roberts, B. W. (2009). Back to the future: Personality and Assessment and personality development. Journal of Research in Personality, 43, 137– 145. doi: 10.1016/j.jrp.2008.12.015
Roberts, B. W., Chernyshenko, O. S., Stark, S., & Goldberg,
L. R. (2005). The structure of conscientiousness: An empirical investigation based on seven major personality questionnaires. Personnel Psychology, 58, 103– 139. doi:
10.1111/j.1744-6570.2005.00301.x
Roberts, B. W., & DelVecchio, W. F. (2000). The rank-order consistency of personality traits from childhood to old age: A quantitative review of longitudinal studies. Psychological Bulletin,
126, 3– 25. doi: 10.1037/0033-2909.126.1.3
Roberts, B. W., Kuncel, N. R., Shiner, R., Caspi, A., & Goldberg, L. R. (2007). The power of personality: The comparative validity of personality traits, socioeconomic
status, and cognitive ability for predicting important life outcomes. Perspectives on Psychological Science, 2, 313– 345. doi:
10.1111/j.1745-6916.2007.00047.x
Roberts, B. W., O’Donnell, M., & Robins, R. W. (2004). Goal and
personality trait development in emerging adulthood. Journal of Personality and Social Psychology, 87, 541– 550. doi:
10.1037/0022-3514.87.4.541
Roberts, B. W., & Robins, R. W. (2000). Broad dispositions, broad
aspirations: The intersection of personality traits and major
110 Wood, Hensler Gardner, & Harms
life goals. Personality and Social Psychology Bulletin, 26, 1284–
1296. doi: 10.1177/0146167200262009
Roberts, B. W., Walton, K. E., & Viechtbauer, W. (2006). Patterns of
mean-level change in personality traits across the life course:
A meta-analysis of longitudinal studies. Psychological Bulletin,
132, 1– 25. doi: 10.1037/0033-2909.132.1.1
Roberts, B. W., & Wood, D. (2006). Personality development in
the context of the neo-socioanalytic model of personality. In
D. K.Mroczek & T. D.Little (Eds.), Handbook of personality development (pp. 11– 39). Mahwah, NJ: Erlbaum.
Roccas, S., Sagiv, L., Schwartz, S. H., & Knafo, A. (2002). The
Big Five personality factors and personal values. Personality and Social Psychology Bulletin, 28, 789– 801. doi:
10.1177/0146167202289008
Saucier, G. (1994). Mini-markers: A brief version of Goldberg’s
unipolar big-five markers. Journal of Personality Assessment, 63,
506– 516. doi: 10.1207/s15327752jpa6303_8
Saucier, G., & Goldberg, L. R. (1996). The language of personality:
Lexical perspectives on the five-factor model. In J. S.Wiggins
(Ed.), The five-factor model of personality: Theoretical perspectives
(pp. 21– 50). New York, NY: Guilford Press.
Saucier, G., & Goldberg, L. R. (2001). Lexical studies of indigenous
personality factors: Premises, products, and prospects. Journal of Personality, 69, 847– 879. doi: 10.1111/1467-6494.696167
Schaller, M., & Murray, D. R. (2010). Infectious diseases and
the evolution of cross-cultural differences. In M.Schaller,
A.Norenzayan, S. J.Heine, T.Yamagishi, & T.Kameda (Eds.),
Evolution, culture, and the human mind (pp. 243– 256). New
York, NY: Psychology Press.
Schwartz, S. H. (1992). Universals in the content and structure of
values: Theoretical advances and empirical tests in 20 countries. Advances in Experimental Social Psychology, 25, 1– 65. doi:
10.1016/S0065-2601(08)60281-6
Skinner, B. F. (1965). Science and human behavior. New York, NY:
Free Press.
Skinner, B. F. (1981). Selection by consequences. Science, 213, 501–
504. doi: 10.1126/science.7244649
Soto, C. J., & John, O. P. (2009). Ten facet scales for the Big Five Inventory: Convergence with NEO PI-R facets, self-peer agreement, and discriminant validity. Journal of Research in Personality, 43, 84– 90. doi: 10.1016/j.jrp.2008.10.002
Soto, C. J., John, O. P., Gosling, S. D., & Potter, J. (2011). Age differences in personality traits from 10 to 65: Big Five domains
and facets in a large cross-sectional sample. Journal of Personality and Social Psychology, 100, 330– 348. doi: 10.1037/a0021717
Specht, J., Egloff, B., & Schmukle, S. C. (2011). Stability and change
of personality across the life course: The impact of age and major life events on mean-level and rank-order stability of the Big
Five. Journal of Personality and Social Psychology, 101, 862– 882.
doi: 10.1037/a0024950
Srivastava, S. (2010). The five-factor model describes the structure of social perceptions. Psychological Inquiry, 21, 69– 75. doi:
10.1080/10478401003648815
Srivastava, S., John, O. P., Gosling, S. D., & Potter, J. (2003). Development of personality in early and middle adulthood: Set
like plaster or persistent change?Journal of Personality and Social
Psychology, 84, 1041– 1053. doi: 10.1037/0022-3514.84.5.1041
Tellegen, A. (1991). Personality traits: Issues of definition, evidence, and assessment. In D.Cicchetti & W. M.Grove (Eds.),
Thinking clearly about psychology: Essays in honor of Paul E.
Meehl, Vol. 2: Personality and psychopathology (pp. 10– 35). Minneapolis, MN: University of Minnesota Press.
in
Psychological Review 122 (2015)
Terracciano, A., & McCrae, R. R. (2012). Why do (some) birds
flock? Causality and the structure of characteristic adaptations. European Journal of Personality, 26, 449– 450. doi: 10.1002/
per.1878
Thorndike, E. L. (1913). The psychology of learning (Vol. 2). New
York, NY: Mason-Henry.
Tolman, E. C. (1938). The determiners of behavior at a choice
point. Psychological Review, 45, 1– 41. doi: 10.1037/h0062733
Tooby, J., & Cosmides, L. (1990). The past explains the present: Emotional adaptations and the structure of ancestral
environments. Ethology and Sociobiology, 11, 375– 424. doi:
10.1016/0162-3095(90)90017-Z
Vaidya, J. G., Gray, E. K., Haig, J., & Watson, D. (2002). On the temporal stability of personality: Evidence for differential stability
and the role of life experiences. Journal of Personality and Social
Psychology, 83, 1469– 1484. doi: 10.1037/0022-3514.83.6.1469
Van Egeren, L. F. (2009). A cybernetic model of global personality traits. Personality and Social Psychology Review, 13, 92– 108.
doi: 10.1177/1088868309334860
Vroom, V. H. (1964). Work and motivation. Oxford, UK: Wiley.
Watson, D., & Clark, L. A. (1997). Extraversion and its positive
emotional core. In R.Hogan, J. A.Johnson, & S. R.Briggs (Eds.),
Handbook of personality psychology (pp. 767– 793). San Diego,
CA: Academic Press. doi: 10.1016/B978-012134645-4/50030-5
Wiggins, J. S. (1997). In defense of traits. In R.Hogan, J. A.Johnson,
& S. R.Briggs (Eds.), Handbook of personality psychology (pp.
95– 115). San Diego, CA: Academic Press. doi: 10.1016/
B978-012134645-4/50005-6
Wood, D. (in press). Testing the lexical hypothesis: Are more
important traits more densely reflected in the English
lexicon?Journal of Personality and Social Psychology.
Wood, D., & Denissen, J. J. A. (in press). Personality stability
and variability across the life course. In N. R.Branscombe &
K.Reynolds (Eds.), Psychology of change: Life contexts, experiences, and identities. New York: Psychology Press.
Wood, D., Harms, P., & Vazire, S. (2010). Perceiver effects as projective tests: What your perceptions of others say about you.
Journal of Personality and Social Psychology, 99, 174– 190. doi:
10.1037/a0019390
Wood, D., Larson, R. W., & Brown, J. R. (2009). How adolescents
come to see themselves as more responsible through participation in youth programs. Child Development, 80, 295– 309. doi:
10.1111/j.1467-8624.2008.01260.x
Wood, D., Nye, C. D., & Saucier, G. (2010). Identification and measurement of a more comprehensive set of person-descriptive
trait markers from the English lexicon. Journal of Research in
Personality, 44, 258– 272. doi: 10.1016/j.jrp.2010.02.003
Wood, D., Tov, W., & Costello, C. (in press). What a ______ thing
to do! Formally characterizing actions by their expected effects. Journal of Personality and Social Psychology.
Wood, D., & Wortman, J. (2012). Trait means and desirabilities as artifactual and real sources of differential stability
of personality traits. Journal of Personality, 80, 665– 701. doi:
10.1111/j.1467-6494.2011.00740.x
Wood, W., & Neal, D. T. (2007). A new look at habits and the
habit-goal interface. Psychological Review, 114, 843– 863. doi:
10.1037/0033-295X.114.4.843
Yang, Y., Read, S. J., Denson, T. F., Xu, Y., Zhang, J., & Pedersen,
W. C. (2014). The key ingredients of personality traits: Situations, behaviors, and explanations. Personality and Social Psychology Bulletin, 40, 79– 91. doi: 10.1177/0146167213505871
Functionalist
and
P r o c e s s A pp r o a c h e s
to
Behavior Explain Trait Covariation Appendix A: Measures of Big Five-Related Traits
Used in Study 1
Assertive: BFI: Has an Assertive personality; IIDL: Bold/assertive
Sociable: BFI: Is Outgoing, sociable; IIDL: Outgoing/sociable
Warm: BFI: can be cold and aloof R, IIDL: kind-hearted/caring;
unfriendly/cold R; affectionate/loving)
Polite: BFI: is sometimes rude to others R; IIDL: rude/inconsiderate R, courteous/polite, pleasant/agreeable
Reliable: BFI: Is a reliable worker; IIDL: Dependable/reliable, undependable/unreliable R
111
Orderly: BFI: Tends to be disorganized; disorganized/messy
Nervous: BFI: Worries a lot; Remains calm in tense situations R;
can Be tense; IIDL: Tense/anxious
Irritable: BFI: Can be moody; IIDL: Crabby/grouchy; angry/hostile; touchy/temperamental
Creative: BFI: Is original, comes up with new ideas; Is inventive;
IIDL: Creative/imaginative
Unconventional: IIDL: Radical/rebellious; traditional/conventional R
Note: Subscript “R” indicates item was reverse-scored.
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 1
Supplemental Materials
How Functionalist and Process Approaches to Behavior Can Explain Trait Covariation
by D. Wood et al., 2014, Psychological Review
http://dx.doi.org/10.1037/a0038423
Supplementary Materials 1 (S1)
Extended Study 1 and Study 2 Methods
Study 1 Method
Part 1: Generating Potential FIs Underlying Big Five-Related Behaviors
A total of 529 Wake Forest University undergraduates over two semesters completed an online
survey in order to earn credit towards a course research participation requirement. Participants ranged in
age from 17 to 23 (M = 18.68; 59% female). Participants were informed that they might be invited to
participate in an interview on the basis of their answers.
Big Five trait assessments. Participants completed the Big Five Inventory (BFI; John &
Srivastava, 1999), Inventory of Individual Differences in the Lexicon (IIDL; Wood, Nye, & Saucier,
2010), and items from the International Personality Item Pool (Goldberg, 1999) identified as highly
associated with Big Five trait levels; many of the IPIP items are provided in Supplementary Materials 2
(S2). Big Five estimates for the IIDL were created by averaging the five items with the highest
correlations with a given Big Five trait reported in Wood, Nye, and Saucier (2010). Big Five scores from
the BFI, IIDL, and IPIP measures for each dimension were then standardized and averaged. The
reliability of these three-scale composites was .95 for Extraversion, .91 for Agreeableness, .91 for
Conscientiousness, .93 for Neuroticism, and .85 for Openness.
Interviews with participants high and low on Big Five dimensions. Participants with the
highest and lowest scores from the three-measure Big Five composites were invited to participate in oneon-one interviews for additional class credit. Each interview lasted approximately 15 to 20 minutes. Five
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 2
to six individuals from each end of each Big Five trait were interviewed, resulting in a total of 52
interviews.
Interviewed participants were asked to describe the extent to which they performed eight to ten
behaviors related to the trait they were selected for, and why. These behaviors were adapted from IPIP
items found to be highly related to Big Five trait scores. For instance, the IPIP items “I start
conversations” and “I don’t talk a lot” were rephrased as questions “Are you typically someone who starts
conversations?” and “Would you say that you talk a lot?” A full listing of the items asked in these
interviews is provided in the Supplementary Materials 2 (S2).
Interviewers were instructed to probe for reasons interviewees performed the behaviors at that
level. In particular, interviewers were instructed to ask participants if there were things (a) that they
liked/disliked about doing the behavior, or that made them seek/avoid doing the behavior, (b) that made it
easy/hard for them to perform the behavior, and (c) about any other aspects of the situation that
influenced their decision to act the way they did. This process continued until either all questions were
asked or 20 minutes had passed.
Reports of others’ high and low Big Five trait levels. All participants who completed the initial
survey in the second semester of data collection (N=229) were asked at the conclusion of the survey to
think of someone they knew who acted in an extremely trait-typical way, and to describe reasons for their
behavior. This was done to elicit additional functions that may not have been provided by participants in
explaining their own behavior.
Each participant was randomly assigned to describe someone they knew who was high or low on
one of the Big Five traits. Descriptions of the desired target were created by using three synonymous
adjectives and a pair of behaviors highly associated with the Big Five dimension. For instance, in the high
extraversion condition, participants were asked “Think of someone you know who is very sociable,
extraverted and outgoing. This is someone who regularly starts conversations with others and who
regularly talks to lots of different people at parties.” Between 21 and 26 individuals were assigned to each
of the 10 conditions (two ends of each Big Five trait); instructions for the remaining traits are in the
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 3
Supplementary Materials 2 (S2).
Participants were then instructed to respond to the following questions: “First, list some instances
in which you recall this person acting in the ways just described.” To elicit reasons for these behaviors,
participants were then asked: “What do you think are some of the reasons that he or she tends to act in
this way?”; “What are some of the things that make it easy for the person to act this way? What are some
of the reasons that make it hard for the person to act in a different way?”; and finally “Put yourself in this
person’s shoes. Why do you think this person wants to act in this way? Why do you think this person does
not want to act in a different way?”
Extraction of reasons for trait-related behavior from interviews and reports. Research
assistants then extracted reasons for trait-related behaviors from the participant interviews and reports of
others’ behavior. Coders were given instructions describing what constituted an appropriate “reason” for
trait-related behavior, which consisted of statements of different types of valuations and goals, abilities,
and effects/expectancies. Two coders listened separately to each recorded interview and copied verbatim
any reasons that the interviewees provided to explain their behavior. Coders then reconciled discrepancies
while listening to the interview a second time together. Finally, each reason was summarized into a short
phrase or sentence. For the free-response survey answers of others’ behaviors, the second author extracted
reasons from the responses provided, making each reason into a one-sentence item. Ultimately, 1,985
reasons were initially extracted across all Big Five traits.
Reduction of reasons for trait-related behavior. Three coders (the second author and two
research assistants) were then provided with instructions to sort this larger set of 1,985 items into a
smaller set of item groups to eliminate redundancies within each Big Five trait. To aid with this task, they
were instructed to first classify each item into one of nine more specific categories: (1) abilities; (2)
behavior-outcome expectancies; (3) situation construals; (4) felt pressures and needs; (5) likes and
dislikes; (6) preferences; (7) values and standards; (8) concerns and worries, and (9) goals. Coders were
then instructed to group similar items while maintaining as many distinctions as possible. After doing this
separately, coders met to form a unified set of distinct reasons for high or low levels of each Big Five
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 4
trait. This was done separately for each Big Five trait, resulting in a list of 633 item groups.
Following this, a group of four coders (the first and second authors and two research assistants)
met again to further reduce redundancies across all Big Five traits. Also at this stage, preference items
were split apart to make separate items involving how much the person liked each object implied in the
preference item (e.g., “I prefer being alone” was separated to “I like being alone” and “I like being with
people”). Following this stage, the list of reasons for Big Five trait-related behavior was further reduced
into a smaller list of 463 distinct reasons.
Part 2: Linking Functionality indicators to Big-Five Related Traits
We continued by exploring how these FIs were empirically associated with variation in
behavioral traits associated with the Big Five.
Participants. A total of 537 Wake Forest University students from an introductory psychology
course completed the items described above via an online survey. Participants were removed if they left
over 20 of the items blank, or if they had no variability in their responses for major sections of the survey
(e.g., answering “2” to every question within a particular subsection). These removals resulted in a final
sample size of 511 participants ranging in age from 17 to 37 years (M = 18.7, 57% female).
Measures of Big-Five Related Behavioral Traits. Participants completed self-ratings of the BFI
(John & Srivastava, 1999) and the IIDL (Wood, Nye, & Saucier, 2010). Using items across these two
inventories, we estimated two distinct behavioral traits within each Big Five trait domain. In constructing
these measures, we excluded any items that concerned self-perceptions of valuations or goals, abilities, or
expectancies (e.g., the BFI Agreeableness item “likes to cooperate with others”), to focus on more clearly
behavioral trait items and self-perceptions. We also attempted to measure traits close to the two major
“subfacets” within each Big Five domain recently described by DeYoung, Quilty, and Peterson (2007)
and Soto and John (2009). Following these considerations, the items used to construct these 10 scales are
given in Appendix A, and alpha values are provided in Table 1.
FIs associated with Big Five-related behaviors. The 463 FIs ultimately generated from Part 1
were adapted into questionnaire statements using four different question-response formats. Items
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 5
pertaining to likes and dislikes were rated under the instruction “How much do you like or dislike the
following things?” with a scale ranging from 1 (Strongly dislike this) to 5 (Strongly like this). Items
pertaining to goals were rated under the instruction “How much do you try or want to do the following
behaviors?” with a scale ranging from 1 (I try very hard to avoid doing this) to 5 (I try very hard to do
this). Items pertaining to abilities were rated under the instruction “How easy or hard do you find doing
the following things when you try to (or feel that you should)?” with a scale ranging from 1 (I find it very
difficult to do this) to 5 (I find it very easy to do this). All remaining items were rated under the general
instruction “How much do you agree with each statement?” with a response scale ranging from 1
(Strongly disagree) to 5 (Strongly agree). The complete inventory is available from the first author.
We then reduced the complete set of 463 items to a more manageable set of approximately 100
items. To identify content most frequently reflected in the inventory and to organize content similarities,
we conducted a hierarchical cluster analysis, using procedures similar to those described by Wood, Nye
and Saucier (2010). We first constructed a dendrogram using the within-group linkage algorithm; in order
to allow antonymous content to be placed on the same cluster, we included all 463 items in their original
form as well as reverse-scored variables of all items, resulting in a cluster analysis of 926 items. We
considered items as forming a cluster if at least two items clustered together in the dendrogram by
showing intercorrelations of a magnitude of .35 or higher. Many of the larger clusters were then broken
into smaller subclusters when there was clear evidence that the subsets of the items reflected different
gradients of meaning. This was indicated more formally by entering the items from the larger clusters into
a factor analysis using principle axis factoring and oblimin rotations, and identifying if there were two or
more groups of items (each consisting of at least two items) which had fairly distinct factor loadings from
one another, generally by having at least two items on each factor with at least .60 loadings and minor
cross-loadings. These procedures resulted in the extraction of 87 clusters.
Finally, we correlated all 463 items with the 61 items of the IIDL. We used this to aid in selecting
one item to represent each of the 87 clusters; items with greater correlations with the IIDL items (either
by having a large maximum correlation, or by having many IIDL items correlated at a level of |r| ≥ .10)
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 6
were given preference. We also examined this matrix to identify additional single items that were not
located on multi-item clusters but that showed large correlations with an IIDL item, or that showed
correlations above an absolute magnitude of .20 with 10 or more IIDL items. These additional
considerations resulted in the identification of an additional 12 items, resulting in a total of 99 FI items.
Study 2 Method
Participants
A total of 700 ESCS participants completed the materials examined here as part of an ongoing
study. Participants ranged in age from 18 to 85 (M = 51.4, 56% female), and were of all levels of
education. See Goldberg (2008) for additional details.
Materials
Saucier Mini-Markers. In this study, we utilized the Saucier Mini-Markers (SMM; Saucier,
1994) as our measure of personality due to its unique advantage among resources collected within the
ESCS sample of being administered multiple times in the form of both self-ratings and peer-ratings. In the
fall of 1998, participants both completed the SMM themselves, and were asked to recruit up to three
people they knew well to describe them on the SMM. Additionally, the participants rated themselves on
the SMM earlier in the summer of 1993, and in the spring of 1995. Consequently, there were up to three
self-ratings and up to three peer-ratings of the SMM that could be used for each participant.
Similar to Study 1, we selected two indicators from each Big Five domain, which were examined
separately. The two trait indicators were selected (1) to measure distinct behavioral traits within the Big
Five domain, and (2) to the extent possible parallel the traits examined in Study 1 (i.e., the traits listed
within Table 1). We also only included items that were positive indicators of the dimension (e.g., for
kindness, the item “kind” would be a positive indicator and “harsh” a negative indicator). This resulted in
the selection of the items bold and extraverted within the domain of Extraversion; kind and cooperative in
the domain of Agreeableness; organized and practical in the domain of Conscientiousness, fretful and
temperamental in the domain of Neuroticism, and creative and philosophical in the domain of Openness.
Scales were formed by aggregating the self-ratings of these traits made in the three different
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 7
administrations; the reliabilities are shown in Table 3 and ranged from .62 to .86. Peer rating scales were
formed by aggregating the 1 to 3 peer ratings obtained in Fall 1998. The intraclass correlations for peer
ratings of the same participant, shown in Table 3, ranged from .13 to .35. Since participants were rated by
an average of 2.5 peers, using the Spearman-Brown prophecy formula we estimated the approximate
reliabilities of the average peer-rated scales as ranging from .27 to .57.
International Personality Item Pool (IPIP). ESCS participants completed up to 2492 distinct
items between Spring 1994 and Spring 2003. These items consisted of relatively short items in which
people described a wide range of behavioral traits, feelings, skills, beliefs, and more abstract selfperceptions. Participants rated about 50 of these items both in Spring 1994 and again in Fall 1995,
allowing for estimation of the test-retest reliability over about a year. For these items, the test-retest
correlations averaged .52, which we use as an approximation of the one-year test-retest reliability or
dependability of the IPIP items.
Identification of functionality indicators (FIs) within the IPIP items. Given the heterogeneity of
content within the IPIP, the first and third authors and a research assistant categorized the IPIP items into
1 of 7 categories. The first six collectively consist of the IPIP items we considered FIs; we list the
categories and some common IPIP item stems: (1) likes/dislikes (e.g., “[Like/dislike]…”, “[Prefer/prefer
not] to…” “Feel [happy/bad] when…”; (2) goals (e.g., “[Want/seek/avoid]…” “[Try/try not] to…”), (3)
values (e.g., “People [should/shouldn’t]…” “Expect others [to/not to]…” “[Allow/let]…” “It is important
[to/not to]…”), (4) contingencies of emotion/attention (e.g., “Am [concerned/not concerned] about…”
“[Pay/don’t pay] attention to…”, “When in [situation], I feel [emotion]”); (5) abilities (e.g.,
“[Can/can’t]…” “Am [easily/not easily]…” “[Know/Don’t know] how to…” “Am [good/bad] at…”); (6)
beliefs/situation perceptions (e.g., “[Believe/do not believe] that…” “[Experience/feel
that]…”“[Know/don’t know] that…”).
Finally, outside of these categories, items could be categorized as concerning (7)
behaviors/identities/reputations, which especially concerned rates of behavior (e.g., “Tend to…”, “When
in [situation], do [response]”), expected rates of behavior (e.g., “Would [probably/never]” and abstract
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 8
trait perceptions. We also placed items in this category that had functional content that was vague or nonspecific (e.g., “Worry about minor things”).
There was relatively consistent categorization of these items into these categories: 1451 of the
total 2413 IPIP items (60%) were placed in the same category by all raters; 785 (33%) were placed into
the same category by two of the three raters, and only 177 (7%) were placed into a different category by
each rater; although frequently all raters categorized such items into one of the first six FI categories. Any
discrepancy beyond universal placement was discussed by the three raters. Of the 2492 items contained
within the IPIP, 1351 (54%) were categorized as perceptions of FIs; these categorizations are available
from the first author upon request.
Identifying functions associated with Big Five-related trait perceptions. We conducted a
multi-step process to identify FIs related to the ten traits examined, and to reduce these to a smaller set
that could be used to represent the diverse functions within the body of the text. Although this procedure
was similar to that used in Study 1, some differences were necessary given the presence of both self- and
peer-ratings of these participants and the much larger item pool used in this sample.
First, as the pattern of correlations between IPIP items and trait perceptions tended to be similar
across both self-ratings and peer-ratings of traits (across the 1351 functional IPIP items, the vector
correlation for how items were associated with self- and peer-ratings ranged from .94 for “extraverted” to
.57 for “practical”), we averaged the item’s correlation with the self-reported trait and with the peerreported trait together. Second, for each of the 10 traits, we found the 100 items that had the highest
absolute correlation with the trait. This resulted in reducing the 1351 functional IPIP items to a smaller set
of the 630 items that were most highly associated with the 10 trait perceptions across self- and peerreports. These items and their reversals were then entered into a cluster analysis, where we pre-specified
the extraction of 75 clusters, again using the within-group average linkage algorithm, and using
correlations as the similarity coefficient. These clusters ranged in size from 24 items, to one cluster with
one item which was not considered further.
For the remaining 74 clusters, we selected a single item to indicate how the cluster was associated
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 9
with trait perceptions. We chose to use single items in order to make the number of items used to
represent the cluster uniform, and so the cluster could be represented by providing the complete item
within Table 4. To help in the selection of an item that indicated how the cluster as a whole related to Big
Five-related trait perceptions, we conducting an inverse factor analysis, where the items within the cluster
were entered as variables and the rows depicted the item’s correlations with the 20 trait perceptions,
considering self- and peer-ratings separately. The items with the highest factor loading within the
associated covariance matrix provided an indication of which items were most representative of the
cluster. We then selected an item from among the highest loading items which seemed to refer
semantically to the content of the cluster and to a specific valuation, ability, and or perceived effect or
situation construal. For instance, the items “Suspect hidden motives in others” and “Don’t care about the
rules” were selected over the items “Distrust others” and “Resist society’s rules,” respectively, due to the
former items referring less ambiguously to FIs rather than to behavioral tendencies.
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 10
Supplementary Materials 2 (S2)
Materials for Generating Reasons for Trait-Related Behavior
1. Interview Questions for Generating Explanations of Big Five Related Behaviors
Original IPIP Item
Extraversion
I start conversations.
I don’t talk a lot.
I am skilled in handling social situations.
I keep in the background.
I talk to lots of different people at parties.
I find it difficult to approach others.
I make friends easily.
I often feel uncomfortable around others.
I warm up quickly to others.
I seem to derive less enjoyment from
interacting with other people than others
do.
Agreeableness
I sympathize with others’ feelings.
I insult people.
I respect others.
I look down on others.
I accept people as they are.
I get back at others.
I find that it takes a lot to make me feel
angry at someone.
I point out others’ mistakes.
I listen to people’s problems.
I tell offensive jokes.
Conscientiousness
I follow through with my plans.
I don’t finish the things that you start.
I usually take care of my responsibilities as
soon as possible.
Interview Item
Are you typically someone that starts conversations?
Would you say that you talk a lot?
How skilled would you say that you are in handling social
situations?
Do you feel that you keep to the background in social
situations, or that you make yourself prominent?
Would you say that you talk to lots of different people at
parties?
How easy do you feel that it is to approach other people?
Would you say that you make friends easily?
How comfortable are you around other people?
Would you say that you are someone who warms up quickly
to others?
Do you feel that you get more or less enjoyment from
interacting with people than others do?
Would you say that you tend to sympathize with other
peoples’ feelings?
Would you say that you are someone who insults people?
How much would you say that you respect other people?
Do you feel that you look down on other people?
Would you say that you tend to accept people as they are?
If someone does you wrong, will you tend to try to get back
at them?
Would you say that it takes a lot to make you angry at
someone?
Would you say that you are someone who points out mistakes
that other people make?
Are you someone who will listen to people’s problems?
Do you tell offensive jokes?
Would you say that you typically follow through with plans
you make?
Would you say that you finish the things that you start?
When you have responsibilities, would you say that you are
someone who takes care of them as soon as you can?
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 11
I find it difficult to organize tasks and
activities.
I complete tasks successfully.
I have difficulty keeping my attention on a
task.
I like to organize things.
I hardly ever finish things on time.
Do you have difficulty organizing tasks and activities?
Are you someone who completes tasks successfully?
Do you have difficulty keeping your attention on tasks?
Are you someone who likes to organize things?
Are you someone who finishes things on time?
Emotional Stability
I am relaxed most of the time.
I get stressed out easily.
I remain calm under pressure.
I panic easily.
I rarely worry.
I am moody.
I am not easily bothered by things.
I am afraid of many things.
Are you usually a relaxed person?
Are you someone who gets stressed out easily?
Are you someone who remains calm under pressure?
Would you say you are someone who panics easily?
Would you say that you are someone who worries a lot?
Are you a moody person?
Are you easily bothered by things?
Would you say that you are afraid of many things?
Openness
I believe in the importance of art.
I seldom notice the emotional aspects of
paintings and pictures.
I have a vivid imagination.
I am not interested in abstract ideas.
I see beauty in things that others might not
notice.
I do not like art.
I enjoy hearing new things.
I have difficulty understanding abstract
ideas.
Do you believe in the importance of art?
Are you someone who notices the emotional aspects of
paintings and pictures?
Do you have a vivid imagination?
Are you someone who is interested in abstract ideas?
Do you think that you see beauty in things that others might
not notice?
Do you like art?
Are you someone who enjoys hearing new things?
Do you have an easy or difficulty time understanding abstract
ideas?
FUNCTIONAL EXPLANATIONS OF TRAIT COVARIATION 12
2. Materials for Generating Explanations of Big Five-Related Behaviors in Others
High Extraversion: Think of someone you know who is very sociable, extraverted, and outgoing. This is
someone who regularly starts conversations with others and who regularly talks to lots of different people
at parties. [26]
Low Extraversion: Think of someone you know who is very reserved, introverted, and shy. This is
someone who regularly keeps in the background in social situations and who regularly has a difficult time
approaching others. [22]
High Agreeableness: Think of someone you know who is very compassionate, agreeable, and kindhearted. This is someone who regularly sympathizes with other people’s feelings and who regularly
listens to other people’s problems.[21]
Low Agreeableness: Think of someone you know who is very inconsiderate, disagreeable, and rude.
This is someone who regularly insults other people and who often offends others. [23]
High Conscientiousness: Think of someone you know who is very dependable, conscientious, and
organized. This is someone who regularly follows through with the plans they make and who regularly
completes task on time. [22]
Low Conscientiousness: Think of someone you know who is very disorganized, unconscientious, and
unreliable. This is someone who regularly starts tasks but doesn’t finish them and who regularly has
trouble keeping his or her attention on a task. [25]
High Emotional Stability: Think of someone you know who is very relaxed, calm, and emotionally
stable. This is someone who regularly remains calm under pressure and who generally is not bothered by
things that could easily upset other people. [22]
Low Emotional Stability: Think of someone you know who is very tense, anxious, and nervous. This is
someone who regularly gets stressed out easily and who gets worried over small things. [23]
High Openness to Experience: Think of someone you know who is very curious, open to new
experiences, and imaginative. This is someone who believes in the importance of art and who frequently
sees beauty in things other people might not notice. [24]
Low Openness to Experience: Think of someone you know who is uninterested in new experiences,
unimaginative, and has fairly narrow interests. This is someone who does not tend to notice the emotional
aspects of art and who regularly has a difficult time understanding abstract ideas. [21]
Note. Number in parentheses provides the number of participants who provided a report for this
instruction set.